How-To Tutorials - Data

Decision tree learners are powerful classifiers that utilize a tree structure to model the relationships among the features and the potential outcomes. This structure earned its name due to the fact that it mirrors the way a literal tree begins at a wide trunk and splits into narrower and narrower branches as it is followed upward. In much the same way, a decision tree classifier uses a structure of branching decisions that channel examples into a final predicted class value. In this article, we demonstrate the implementation of decision tree using C5.0 algorithm in R. This article is taken from the book, Machine Learning with R, Fourth Edition written by Brett Lantz. This 10th Anniversary Edition of the classic R data science book is updated to R 4.0.0 with newer and better libraries. This book features several new chapters that reflect the progress of machine learning in the last few years and help you build your data science skills and tackle more challenging problems There are numerous implementations of decision trees, but the most well-known is the C5.0 algorithm. This algorithm was developed by computer scientist J. Ross Quinlan as an improved version of his prior algorithm, C4.5 (C4.5 itself is an improvement over his Iterative Dichotomiser 3 (ID3) algorithm). Although Quinlan markets C5.0 to commercial clients (see http://www.rulequest.com/ for details), the source code for a single-threaded version of the algorithm was made public, and has therefore been incorporated into programs such as R. The C5.0 decision tree algorithm The C5.0 algorithm has become the industry standard for producing decision trees because it does well for most types of problems directly out of the box. Compared to other advanced machine learning models, the decision trees built by C5.0 generally perform nearly as well but are much easier to understand and deploy. Additionally, as shown in the following table, the algorithm's weaknesses are relatively minor and can be largely avoided. Strengths An all-purpose classifier that does well on many types of problems. Highly automatic learning process, which can handle numeric or nominal features, as well as missing data. Excludes unimportant features. Can be used on both small and large datasets. Results in a model that can be interpreted without a mathematical background (for relatively small trees). More efficient than other complex models. Weaknesses Decision tree models are often biased toward splits on features having a large number of levels. It is easy to overfit or underfit the model. Can have trouble modeling some relationships due to reliance on axis-parallel splits. Small changes in training data can result in large changes to decision logic. Large trees can be difficult to interpret and the decisions they make may seem counterintuitive. To keep things simple, our earlier decision tree example ignored the mathematics involved with how a machine would employ a divide and conquer strategy. Let's explore this in more detail to examine how this heuristic works in practice. Choosing the best split The first challenge that a decision tree will face is to identify which feature to split upon. In the previous example, we looked for a way to split the data such that the resulting partitions contained examples primarily of a single class. The degree to which a subset of examples contains only a single class is known as purity, and any subset composed of only a single class is called pure. There are various measurements of purity that can be used to identify the best decision tree splitting candidate. C5.0 uses entropy, a concept borrowed from information theory that quantifies the randomness, or disorder, within a set of class values. Sets with high entropy are very diverse and provide little information about other items that may also belong in the set, as there is no apparent commonality. The decision tree hopes to find splits that reduce entropy, ultimately increasing homogeneity within the groups. Typically, entropy is measured in bits. If there are only two possible classes, entropy values can range from 0 to 1. For n classes, entropy ranges from 0 to log2(n). In each case, the minimum value indicates that the sample is completely homogenous, while the maximum value indicates that the data are as diverse as possible, and no group has even a small plurality. In mathematical notion, entropy is specified as: In this formula, for a given segment of data (S), the term c refers to the number of class levels, and pi  refers to the proportion of values falling into class level i. For example, suppose we have a partition of data with two classes: red (60 percent) and white (40 percent). We can calculate the entropy as: > -0.60 * log2(0.60) - 0.40 * log2(0.40) [1] 0.9709506 We can visualize the entropy for all possible two-class arrangements. If we know the proportion of examples in one class is x, then the proportion in the other class is (1 – x). Using the curve() function, we can then plot the entropy for all possible values of x: > curve(-x * log2(x) - (1 - x) * log2(1 - x),     col = "red", xlab = "x", ylab = "Entropy", lwd = 4) This results in the following figure: The total entropy as the proportion of one class varies in a two-class outcome As illustrated by the peak in entropy at x = 0.50, a 50-50 split results in the maximum entropy. As one class increasingly dominates the other, the entropy reduces to zero. To use entropy to determine the optimal feature to split upon, the algorithm calculates the change in homogeneity that would result from a split on each possible feature, a measure known as information gain. The information gain for a feature F is calculated as the difference between the entropy in the segment before the split (S1) and the partitions resulting from the split (S2): One complication is that after a split, the data is divided into more than one partition. Therefore, the function to calculate Entropy(S2) needs to consider the total entropy across all of the partitions. It does this by weighting each partition's entropy according to the proportion of all records falling into that partition. This can be stated in a formula as: In simple terms, the total entropy resulting from a split is the sum of entropy of each of the n partitions weighted by the proportion of examples falling in the partition (wi). The higher the information gain, the better a feature is at creating homogeneous groups after a split on that feature. If the information gain is zero, there is no reduction in entropy for splitting on this feature. On the other hand, the maximum information gain is equal to the entropy prior to the split. This would imply the entropy after the split is zero, which means that the split results in completely homogeneous groups. The previous formulas assume nominal features, but decision trees use information gain for splitting on numeric features as well. To do so, a common practice is to test various splits that divide the values into groups greater than or less than a threshold. This reduces the numeric feature into a two-level categorical feature that allows information gain to be calculated as usual. The numeric cut point yielding the largest information gain is chosen for the split. Note: Though it is used by C5.0, information gain is not the only splitting criterion that can be used to build decision trees. Other commonly used criteria are Gini index, chi-squared statistic, and gain ratio. For a review of these (and many more) criteria, refer to An Empirical Comparison of Selection Measures for Decision-Tree Induction, Mingers, J, Machine Learning, 1989, Vol. 3, pp. 319-342. Pruning the decision tree As mentioned earlier, a decision tree can continue to grow indefinitely, choosing splitting features and dividing into smaller and smaller partitions until each example is perfectly classified or the algorithm runs out of features to split on. However, if the tree grows overly large, many of the decisions it makes will be overly specific and the model will be overfitted to the training data. The process of pruning a decision tree involves reducing its size such that it generalizes better to unseen data. One solution to this problem is to stop the tree from growing once it reaches a certain number of decisions or when the decision nodes contain only a small number of examples. This is called early stopping or prepruning the decision tree. As the tree avoids doing needless work, this is an appealing strategy. However, one downside to this approach is that there is no way to know whether the tree will miss subtle but important patterns that it would have learned had it grown to a larger size. An alternative, called post-pruning, involves growing a tree that is intentionally too large and pruning leaf nodes to reduce the size of the tree to a more appropriate level. This is often a more effective approach than prepruning because it is quite difficult to determine the optimal depth of a decision tree without growing it first. Pruning the tree later on allows the algorithm to be certain that all of the important data structures were discovered. Note: The implementation details of pruning operations are very technical and beyond the scope of this book. For a comparison of some of the available methods, see A Comparative Analysis of Methods for Pruning Decision Trees, Esposito, F, Malerba, D, Semeraro, G, IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997, Vol. 19, pp. 476-491. One of the benefits of the C5.0 algorithm is that it is opinionated about pruning—it takes care of many of the decisions automatically using fairly reasonable defaults. Its overall strategy is to post-prune the tree. It first grows a large tree that overfits the training data. Later, the nodes and branches that have little effect on the classification errors are removed. In some cases, entire branches are moved further up the tree or replaced by simpler decisions. These processes of grafting branches are known as subtree raising and subtree replacement, respectively. Getting the right balance of overfitting and underfitting is a bit of an art, but if model accuracy is vital, it may be worth investing some time with various pruning options to see if it improves the test dataset performance. To summarize , decision trees are widely used due to their high accuracy and ability to formulate a statistical model in plain language.  Here, we looked at a highly popular and easily configurable decision tree algorithm C5.0. The major strength of the C5.0 algorithm over other decision tree implementations is that it is very easy to adjust the training options. Harness the power of R to build flexible, effective, and transparent machine learning models with Brett Lantz’s latest book Machine Learning with R, Fourth Edition. Dr.Brandon explains Decision Trees to Jon Building a classification system with Decision Trees in Apache Spark 2.0 Implementing Decision Trees

Behind the heels of Stanford’s HCAI Institute ( which, mind you, received public backlash for non-representative faculty makeup). Amazon is collaborating with the National Science Foundation (NSF) to develop systems based on fairness in AI. The company will be investing $10M each in artificial intelligence research grants over a three-year period. The official announcement was made by Prem Natarajan, VP of natural understanding in the Alexa AI group, who wrote in a blog post “With the increasing use of AI in everyday life, fairness in artificial intelligence is a topic of increasing importance across academia, government, and industry. Here at Amazon, the fairness of the machine learning systems we build to support our businesses is critical to establishing and maintaining our customers’ trust.” Per the blog post, Amazon will be collaborating with NSF to build trustworthy AI systems to address modern challenges. They will explore topics of transparency, explainability, accountability, potential adverse biases and effects, mitigation strategies, validation of fairness, and considerations of inclusivity. Proposals will be accepted from March 26 until May 10, to result in new open source tools, publicly available data sets, and publications. The two organizations plan to continue the program with calls for additional proposals in 2020 and 2021. There will be 6 to 9 awards of type Standard Grant or Continuing Grant. The award size will be $750,000 - up to a maximum of $1,250,000 for periods of up to 3 years. The anticipated funding amount is $7,600,000. “We are excited to announce this new collaboration with Amazon to fund research focused on fairness in AI,” said Jim Kurose, NSF's head for Computer and Information Science and Engineering. “This program will support research related to the development and implementation of trustworthy AI systems that incorporate transparency, fairness, and accountability into the design from the beginning.” The insidious nexus of private funding in public research: What does Amazon gain from collab with NSF? Amazon’s foray into fairness system looks more of a publicity stunt than eliminating AI bias. For starters, Amazon said that they will not be making the award determinations for this project. NSF would solely be awarding in accordance with its merit review process. However, Amazon said that Amazon researchers may be involved with the projects as an advisor only at the request of an awardee, or of NSF with the awardee's consent. As advisors, Amazon may host student interns who wish to gain further industry experience, which seems a bit dicey. Amazon will also not participate in the review process or receive proposal information. NSF will only be sharing with Amazon summary-level information that is necessary to evaluate the program, specifically the number of proposal submissions, number of submitting organizations, and numbers rated across various review categories. There was also the question of who exactly is funding since VII.B section of the proposal states: "Individual awards selected for joint funding by NSF and Amazon will be   funded through separate NSF and Amazon funding instruments." https://twitter.com/nniiicc/status/1110335108634951680 https://twitter.com/nniiicc/status/1110335004989521920 Nic Weber, the author of the above tweets and Assistant Professor at UW iSchool, also raises another important question: “Why does Amazon get to put its logo on a national solicitation (for a paltry $7.6 million dollars in basic research) when it profits in the multi-billions off of AI that is demonstrably unfair and harmful.” Twitter was abundant with tweets from those in working tech questioning Amazon’s collaboration. https://twitter.com/mer__edith/status/1110560653872373760 https://twitter.com/patrickshafto/status/1110748217887649793 https://twitter.com/smunson/status/1110657292549029888 https://twitter.com/haldaume3/status/1110697325251448833 Amazon has already been under the fire due to its controversial decisions in the recent past. In June last year, when the US Immigration and Customs Enforcement agency (ICE) began separating migrant children from their parents, Amazon came under fire as one of the tech companies that aided ICE with the software required to do so. Amazon has also faced constant criticisms since the news came that Amazon had sold its facial recognition product Rekognition to a number of law enforcement agencies in the U.S. in the first half of 2018. Amazon is also under backlash after a study by the Massachusetts Institute of Technology in January, found Amazon Rekognition incapable of reliably determining the sex of female and darker-skinned faces in certain scenarios. Amazon is yet to fix this AI-bias anomaly, and yet it has now started a new collaboration with NSF that ironically focusses on building bias-free AI systems. Amazon’s Ring (a smart doorbell company) also came under public scrutiny in January, after it gave access to its employees to watch live footage from cameras of the customers. In other news, yesterday, Google also formed an external AI advisory council to help advance the responsible development of AI. More details here. Amazon won’t be opening its HQ2 in New York due to public protests Amazon admits that facial recognition technology needs to be regulated Amazon’s Ring gave access to its employees to watch live footage of the customers, The Intercept reports

Yesterday, March 21, four versions of Wikipedia, German, Danish, Czech, and Slovak were blacked off as a move to oppose the recent EU Copyright Directive, which will be up for voting on Tuesday, March 26. These long-awaited updates to the copyright law include “important wins for the open community in the current text”, the Wikimedia foundation reports. However, “the inclusion of Articles 11 and 13 will harm the way people find and share information online”, Wikimedia further states. However, the major opposition is towards the controversial Article 13. Article 11 states that if a text contains more than a snippet from an article, it must be licensed and paid for by whoever quotes the text. “While each country can define "snippet" however it wants, the Directive does not stop countries from making laws that pass using as little as three words from a news story”, the Electronic Frontier Foundation mentions. Article 13 is, however, the most controversial and is all set to restructure how copyright works on the web. As of now, in order to take down content that is subject to copyright infringement, the rights holder just have to send a ‘takedown notice’. However, with Article 13 in place, there will be no protection for online services and also “relieves rights-holders of the need to check the Internet for infringement and send out notices. Instead, it says that online platforms have a duty to ensure that none of their users infringe copyright.” According to The Next Web, “To make people understand how serious the effects of the Copyright Reform will be if it’s passed, Reddit and Wikipedia will hinder access to their sites in the EU to mimic the effects of the directive.” Both Article 11 and 13 were reintroduced under the leadership of German Member of the European Parliament (MEP) Axel Voss. However, these had already been discarded as unworkable after expert advice. “Voss's insistence that Articles 11 and 13 be included in the final Directive has been a flashpoint for public anger, drawing criticism from the world's top technical, copyright, journalistic, and human rights experts and organizations”, the Electronic Frontier Foundation reports. “Critics say the politicians behind the legislation do not understand the breadth of the laws they are proposing, and that the directive, if implemented, will harm free expression online”, The Verge reports. Platforms such as Tumblr, YouTube, and many others, that host user-generated content will be under the radar if Article 13 is passed and will be legally responsible if the users upload copyrighted content. According to The Verge, “The only way to stop these uploads, say critics, will be to scan content before its uploaded, leading to the creation of filters that will likely be error-prone and abused by copyright trolls.” Many have protested against Article 13 in recent weeks. In Germany, about 3,500 people took out a rally in Berlin as a protest against the new copyright plans. Also, a petition ‘Save the Internet’ has already gathered more than five million signatures. Reddit has also taken an action against the Copyright Directive by flashing a simulated error message citing failure when Reddit desktop users in EU countries attempt to make a top-level post on Reddit. According to Reddit, “This experience, meant to mimic the automated filters that users would encounter should the Directive pass, will last through March 23rd, when IRL demonstrations are planned across Europe.” Julia Reda, a member of the European Parliament from Germany, in her blog post mentions, “For two years we’ve debated different drafts and versions of the controversial Articles 11 and 13. Now, there is no more ambiguity: This law will fundamentally change the internet as we know it – if it is adopted in the upcoming final vote. But we can still prevent that!” United Nations’ free-speech rapporteur, David Kaye, said, “Europe has a responsibility to modernize its copyright law to address the challenges of the digital age. But this should not be done at the expense of the freedom of expression that Europeans enjoy today… Article 13 of the proposed Directive appears destined to drive internet platforms toward monitoring and restriction of user-generated content even at the point of upload. Such sweeping pressure for pre-publication filtering is neither a necessary nor proportionate response to copyright infringement online.” A user on HackerNews writes, “I hope they win and that Article 11 and 13 will be removed. I think this is an important moment in the birth of EU democracy because it feels to me that one of the first times, there is a big public discussion about an issue and the people at the center aren't national politicians like Merkel or Macron but EU MEPs, namely Voss vs Reda. The EU has rightfully been criticized of not being democratic enough, and this discussion feels like it's very much democratic.” https://twitter.com/Wikipedia/status/1108595296068501504 Five EU countries oppose the EU copyright directive Reddit’s 2018 Transparency report includes copyright removals, restorations, and more! Drafts of Article 13 and the EU Copyright Directive have been finalized

Last Friday’s uncontrolled spread of horrific videos on the Christchurch mosque attack and a propaganda coup for espousing hateful ideologies raised questions about social media. The tech companies scrambled to take action on time due to the speed and volume of content which was uploaded, reuploaded and shared by the users worldwide. In Washington and Silicon Valley, the incident crystallized growing concerns about the extent to which government and market forces have failed to check the power of social media. The failure highlighted the social media companies struggle to police content that are massively lucrative and persistently vulnerable to outside manipulation despite years of promises to do better. After the white supremacist live-streamed the attack and uploaded the video to Facebook, Twitter, YouTube, and other platforms across the internet. These tech companies faced back lashes from the media and internet users worldwide, to an extent where they were regarded as complicit in promoting white supremacism too. In response to the backlash, Google and Facebook provides status report on what they went through when the video was reported, the kind of challenges they faced and what are the next steps to combat such incidents in future. Google’s report so far... Google in an email to Motherboard says it employs 10,000 people across to moderate the company’s platforms and products. They also described a process they would follow when a user reports a piece of potentially violating content—such as the attack video; which is The user flagged report will go to a human moderator to assess. The moderator is instructed to flag all pieces of content related to the attack as “Terrorist Content,” including full-length or sections of the manifesto. Because of the document’s length the email tells moderators not to spend an extensive amount of time trying to confirm whether a piece of content does contain part of the manifesto. Instead, if the moderator is unsure, they should err on the side of caution and still label the content as “Terrorist Content,” which will then be reviewed by a second moderator. The second moderator is told to take time to verify that it is a piece of the manifesto, and appropriately mark the content as terrorism no matter how long or short the section may be. Moderators are told to mark the manifesto or video as terrorism content unless there is an Educational, Documentary, Scientific, or Artistic (EDSA) context to it. Further Google adds that they want to preserve journalistic or educational coverage of the event, but does not want to allow the video or manifesto itself to spread throughout the company’s services without additional context. Google at some point had taken the unusual step of automatically rejecting any footage of violence from the attack video, cutting out the process of a human determining the context of the clip. If, say, a news organization was impacted by this change, the outlet could appeal the decision, Google commented. “We made the call to basically err on the side of machine intelligence, as opposed to waiting for human review,” YouTube’s Product Officer Neal Mohan told the Washington Post in an article published Monday. Google also tweaked the search function to show results from authoritative news sources. It suspended the ability to search for clips by upload date, making it harder for people to find copies of the attack footage. "Since Friday’s horrific tragedy, we’ve removed tens of thousands of videos and terminated hundreds of accounts created to promote or glorify the shooter," a YouTube spokesperson said. “Our teams are continuing to work around the clock to prevent violent and graphic content from spreading, we know there is much more work to do,” the statement added. Facebook’s update so far... Facebook on Wednesday also shared an update on how they have been working with the New Zealand Police to support their investigation. It provided additional information on how their products were used to circulate videos and how they plan to improve them. So far Facebook has provided the following information: The video was viewed fewer than 200 times during the live broadcast. No users reported the video during the live broadcast. Including the views during the live broadcast, the video was viewed about 4,000 times in total before being removed from Facebook. Before Facebook was alerted to the video, a user on 8chan posted a link to a copy of the video on a file-sharing site. The first user report on the original video came in 29 minutes after the video started, and 12 minutes after the live broadcast ended. In the first 24 hours, Facebook removed more than 1.2 million videos of the attack at upload, which were therefore prevented from being seen on our services. Approximately 300,000 additional copies were removed after they were posted. As there were questions asked to Facebook about why artificial intelligence (AI) didn’t detect the video automatically. Facebook says AI has made massive progress over the years to proactively detect the vast majority of the content it can remove. But it’s not perfect. “To achieve that we will need to provide our systems with large volumes of data of this specific kind of content, something which is difficult as these events are thankfully rare.” says Guy Rosen VP Product Management at Facebook. Guy further adds, “AI is an incredibly important part of our fight against terrorist content on our platforms, and while its effectiveness continues to improve, it is never going to be perfect. People will continue to be part of the equation, whether it’s the people on our team who review content, or people who use our services and report content to us. That’s why last year Facebook more than doubled the number of people working on safety and security to over 30,000 people, including about 15,000 content reviewers to report content that they find disturbing.” Facebook further plans to: Improve the image and video matching technology so that they can stop the spread of viral videos of such nature, regardless of how they were originally produced. React faster to this kind of content on a live streamed video. Continue to combat hate speech of all kinds on their platform. Expand industry collaboration through the Global Internet Forum to Counter Terrorism (GIFCT). Challenges Google and Facebook faced to report the video content According to Motherboard, Google saw an unprecedented number of attempts to post footage from the attack, sometimes as fast as a piece of content per second. But the challenge they faced was to block access to the killer’s so-called manifesto, a 74-page document that spouted racist views and explicit calls for violence. Google described the difficulties of moderating the manifesto, pointing to its length and the issue of users sharing the snippets of the manifesto that Google’s content moderators may not immediately recognise. “The manifesto will be particularly challenging to enforce against given the length of the document and that you may see various segments of various lengths within the content you are reviewing,” says Google. A source with knowledge of Google’s strategy for moderating the New Zealand attack material said this can complicate moderation efforts because some outlets did use parts of the video and manifesto. UK newspaper The Daily Mail let readers download the terrorist’s manifesto directly from the paper’s own website, and Sky News Australia aired parts of the attack footage, BuzzFeed News reported. On the other hand Facebook faces a challenge to automatically discern such content from visually similar, innocuous content. For example if thousands of videos from live-streamed video games are flagged by the systems, reviewers could miss the important real-world videos where they could alert first responders to get help on the ground. Another challenge for Facebook is similar to what Google faces, which is the proliferation of many different variants of videos makes it difficult for the image and video matching technology to prevent spreading further. Facebook found that a core community of bad actors working together to continually re-upload edited versions of the video in ways designed to defeat their detection. Second, a broader set of people distributed the video and unintentionally made it harder to match copies. Websites and pages, eager to get attention from people seeking out the video, re-cut and re-recorded the video into various formats. In total, Facebook found and blocked over 800 visually-distinct variants of the video that were circulating. Both companies seem to be working hard to improve their products and gain user’s trust and confidence back. How social media enabled and amplified the Christchurch terrorist attack Google to be the founding member of CDF (Continuous Delivery Foundation) Google announces the stable release of Android Jetpack Navigation

The recently held Open Source Initiative’s 2019 Board elections elected six Board of Directors to its eleven-person Board. Two were elected from the affiliate membership, and four from the individual membership. If it wasn’t incredible enough that many women ran for the seats,  they have won all the seats! The six seats include two returning directors, Carol Smith and Molly de Blanc; and three new directors Pamela Chestek, Elana Hashman, and Hong Phuc Dang. Pamela Chestek (nominated by The Document Foundation) and Molly de Blanc (nominated by the Debian Project) captured the most votes from OSI Affiliate Members. The last seat is a tie between Christine Hall and Mariatta Wijaya and hence a runoff election will be required to identify the final OSI Board Director. The run off election started yesterday, March 18th (opening at 12:00 a.m. / 00:00) and will end on Monday, March 25th (closing at 12:00 a.m. / 00:00). Mariatta Wijaya, a core Python developer and a platform engineer at Zapier, told in a statement to Business Insider that she found not all open source projects were as welcoming, especially to women. That's one reason why she's running for the board of the Open Source Initiative, an influential organization that promotes and protects open source software communities. Wijaya also said, "I really want to see better diversity across the people who contribute to open source. Not just the users, the creators of open source. I would love to see that diversity improve. I would like to see a better representation. I did find it a barrier initially, not seeing more people who look like me in this space, and I felt like an outsider." A person discussed six female candidates in misogynistic language on Slashdot, which is a tech-focussed social news website. The post also then labeled each woman with how much of a "threat" they were. Slashdot immediately took down this post “shortly afterward the OSI started seeing inappropriate comments posted on its website”. https://twitter.com/alicegoldfuss/status/1102609189342371840 Molly de Blanc and Patrick Masson said this was the first time they saw such type of harassment of female OSI board candidates. They also said that such harassments in open source are not uncommon. Joshua R. Simmons, an Open source advocate, and web developer tweeted, “women winning 100% of the open seats in an election that drew attention from a cadre of horrible misogynists” https://twitter.com/joshsimmons/status/1107303020293832704 OSI President, Simon Phipps said that the OSI committee is “thrilled the electorate has picked an all-female cohort to the new Board” https://twitter.com/webmink/status/1107367907825274886 To know more about these elections in detail, head over to the OSI official blog post. UPDATED: In the previous draft, Pamela Chestek who was listed as returning board member, is a new board member; and Carol Smith who was listed as a new board member, is a returning member. #GoogleWalkout demanded a ‘truly equitable culture for everyone’; Pichai shares a “comprehensive” plan for employees to safely report sexual harassment MongoDB withdraws controversial Server Side Public License from the Open Source Initiative’s approval process Google’s pay equity analysis finds men, not women, are underpaid; critics call out design flaws in the analysis  

The recent horrifying terrorist attack in New Zealand has cast new blame on how technology platforms police content. There are now questions about whether global internet services are designed to work this way? And if online viral hate is uncontainable? Fifty one people so far have been reported to be dead and 50 more injured after the terrorist attacks on two New Zealand mosques on Friday. The victims included children as young as 3 and 4 years old, and elderly men and women. The alleged shooter is identified as a 28 year old Australian man named Brenton Tarrant. Brenton announced the attack on the anonymous-troll message board 8chan. There, he posted images of the weapons days before the attack, and made an announcement an hour before the shooting. On 8chan, Facebook and Twitter, he also posted links to a 74-page manifesto, titled “The Great Replacement,” blaming immigration for the displacement of whites in Oceania and elsewhere. The manifesto cites “white genocide” as a motive for the attack, and calls for “a future for white children” as its goal. Further he live-streamed the attacks on Facebook, YouTube; and posted a link to the stream on 8chan. It’s terrifying and disgusting, especially when 8chan is one of the sites where disaffected internet misfits create memes and other messages to provoke dismay and sow chaos among people. “8chan became the new digital home for some of the most offensive people on the internet, people who really believe in white supremacy and the inferiority of women,” Ethan Chiel wrote. “It’s time to stop shitposting,” the alleged shooter’s 8chan post reads, “and time to make a real-life effort post.” Many of the responses, anonymous by 8chan’s nature, celebrate the attack, with some posting congratulatory Nazi memes. A few seem to decry it, just for logistical quibbles. And others lament that the whole affair might destroy the site, a concern that betrays its users’ priorities. Social media encourages performance crime The use of social media technology and livestreaming marks the attack as different from many other terrorist incidents. It is a form of violent “performance crime”. That is, the video streaming is a central component of the violence itself, it’s not somehow incidental to the crime, or a trophy for the perpetrator to re-watch later. In the past, terrorism functioned according to what has been called the “theatre of terror”, which required the media to report on the spectacle of violence created by the group. Nowadays with social media in our hands it's much easier for someone to both create the spectacle of horrific violence and distribute it widely by themselves. There is a tragic and recent history of performance crime videos that use live streaming and social media video services as part of their tactics. In 2017, for example, the sickening murder video of an elderly man in Ohio was uploaded to Facebook, and the torture of a man with disabilities in Chicago was live streamed. In 2015, the murder of two journalists was simultaneously broadcast on-air, and live streamed. Tech companies on the radar Social-media companies scrambled to take action as the news—and the video—of the attack spread. Facebook finally managed to pull down Tarrant’s profiles and the video, but only after New Zealand police brought the live-stream to the company’s attention. It has been working "around the clock" to remove videos of the incident shared on its platform. In a statement posted to Twitter on Sunday, the tech company said that within 24 hours of Friday’s shooting it had removed 1.5 million videos of the attack from its platform globally. YouTube said it had also removed an “unprecedented volume” of videos of the shooting. Twitter also suspended Tarrant’s account, where he had posted links to the manifesto from several file-sharing sites. The chaotic aftermath mostly took place while many North Americans slept unaware, waking up to the news and its associated confusion. By morning on the East Coast, news outlets had already weighed in on whether technology companies might be partly to blame for catastrophes such as the New Zealand massacre because they have failed to catch offensive content before it spreads. One of the tweets say Google, Twitter and Facebook made a choice to not use tools available to them to stop white supremacist terrorism. https://twitter.com/samswey/status/1107055372949286912 Countries like Germany and France already have a law in place that demands social media sites move quickly to remove hate speech, fake news and illegal material. Sites that do not remove "obviously illegal" posts could face fines of up to 50m euro (£44.3m). In the wake of the attack, a consortium of New Zealand’s major companies has pledged to pull their advertising from Facebook. In a joint statement, the Association of New Zealand Advertisers (ANZA) and the Commercial Communications Council asked domestic companies to think about where “their advertising dollars are spent, and carefully consider, with their agency partners, where their ads appear.” They added, “We challenge Facebook and other platform owners to immediately take steps to effectively moderate hate content before another tragedy can be streamed online.” Additionally internet service providers like Vodafone, Spark and Vocus in New Zealand are blocking access to websites that do not respond or refuse to comply to requests to remove reuploads of the shooter’s original live stream. The free speech vs safety debate puts social media platforms in the crosshairs Tech Companies are facing new questions on content moderation following the New Zealand attack. The shooter posted a link to the live stream, and soon after he was apprehended, reuploads were found on other platforms like YouTube and Twitter. “Tech companies basically don’t see this as a priority,” the counter-extremism policy adviser Lucinda Creighton commented. “They say this is terrible, but what they’re not doing is preventing this from reappearing.” Others affirmed the importance of quelling the spread of the manifesto, video, and related materials, for fear of producing copycats, or of at least furthering radicalization among those who would be receptive to the message. The circulation of ideas might have motivated the shooter as much as, or even more than, ethnic violence. As Charlie Warzel wrote at The New York Times, the New Zealand massacre seems to have been made to go viral. Tarrant teased his intentions and preparations on 8chan. When the time came to carry out the act, he provided a trove of resources for his anonymous members, scattered to the winds of mirror sites and repositories. Once the live-stream started, one of the 8chan user posted “capped for posterity” on Tarrant’s thread, meaning that he had downloaded the stream’s video for archival and, presumably, future upload to other services, such as Reddit or 4chan, where other like-minded trolls or radicals would ensure the images spread even further. As Warzel put it, “Platforms like Facebook, Twitter, and YouTube … were no match for the speed of their users.” The internet is a Pandora’s box that never had a lid. Camouflaging stories is easy but companies trying hard in building AI to catch it Last year, Mark Zuckerberg defended himself and Facebook before Congress against myriad failures, which included Russian operatives disrupting American elections and permitting illegal housing ads that discriminate by race. Mark Zuckerberg repeatedly invoked artificial intelligence as a solution for the problems his and other global internet companies have created. There’s just too much content for human moderators to process, even when pressed hard to do so under poor working conditions. The answer, Zuckerberg has argued, is to train AI to do the work for them. But that technique has proved insufficient. That’s because detecting and scrubbing undesirable content automatically is extremely difficult. False positives enrage earnest users or foment conspiracy theories among paranoid ones, thanks to the black-box nature of computer systems. Worse, given a pool of billions of users, the clever ones will always find ways to trick any computer system, for example, by slightly modifying images or videos in order to make them appear different to the computer but identical to human eyes. 8chan, as it happens, is largely populated by computer-savvy people who have self-organized to perpetrate exactly those kinds of tricks. The primary sources of content are only part of the problem. Long after the deed, YouTube users have bolstered conspiracy theories about murders, successfully replacing truth with lies among broad populations of users who might not even know they are being deceived. Even stock-photo providers are licensing stills from the New Zealand shooter’s video; a Reuters image that shows the perpetrator wielding his rifle as he enters the mosque is simply credited, “Social media.” Interpreting real motives is difficult on social The video is just the tip of the iceberg. Many smaller and less obviously inflamed messages have no hope of being found, isolated, and removed by technology services. The shooter praised Donald Trump as a “symbol of renewed white identity” and incited the conservative commentator Candace Owens, who took the bait on Twitter in a post that got retweeted thousands of times by the morning after the attack. The shooter’s forum posts and video are littered with memes and inside references that bear special meaning within certain communities on 8chan, 4chan, Reddit, and other corners of the internet, offering tempting receptors for consumption and further spread. Perhaps worst of all, the forum posts, the manifesto, and even the shooting itself might not have been carried out with the purpose that a literal read of their contents suggests. At the first glance, it seems impossible to deny that this terrorist act was motivated by white-extremist hatred, an animosity that authorities like the FBI expert and the Facebook officials would want to snuff out before it spreads. But 8chan is notorious for users with an ironic and rude behaviour under the shades of anonymity.They use humor, memes and urban slang to promote chaos and divisive rhetoric. As the internet separates images from context and action from intention, and then spreads those messages quickly among billions of people scattered all around the globe. That structure makes it impossible to even know what individuals like Tarrant “really mean” by their words and actions. As it spreads, social-media content neuters earnest purpose entirely, putting it on the same level as anarchic randomness. What a message means collapses into how it gets used and interpreted. For 8chan trolls, any ideology might be as good as any other, so long as it produces chaos. We all have a role to play It’s easy to say that technology companies can do better. They can, and they should. But ultimately, content moderation is not the solution by itself. The problem is the media ecosystem they have created. The only surprise is that anyone would still be surprised that social media produce this tragic abyss, for this is what social media are supposed to do, what they were designed to do: spread the images and messages that accelerate interest and invoke raw emotions, without check, and absent concern for their consequences. We hope that social media companies get better at filtering out violent content and explore alternative business models, and governments think critically about cyber laws that protect both people and speech. But until they do we should reflect on our own behavior too. As news outlets, we shape the narrative through our informed perspectives which makes it imperative to publish legitimate & authentic content. Let’s as users too make a choice of liking and sharing content on social platforms. Let’s consider how our activities could contribute to an overall spectacle society that might inspire future perpetrator-produced videos of such gruesome crime – and act accordingly. In this era of social spectacle, we all have a role to play in ensuring that terrorists aren’t rewarded for their crimes with our clicks and shares. The Indian government proposes to censor social media content and monitor WhatsApp messages Virality of fake news on social media: Are weaponized AI bots to blame, questions Destin Sandlin Mastodon 2.7, a decentralized alternative to social media silos, is now out!

In a hearing on March 12, the sub-committee on emerging threats and capabilities received testimonies on Artificial Intelligence Initiatives within the Department of Defense(DoD). The panel included Peter Highnam, Deputy Director of the Defense Advanced Research Projects Agency; Michael Brown, DoD Defense Innovation Unit Director; and Lieutenant General John Shanahan, director of the Joint Artificial Intelligence Center (JAIC). The panel broadly testified to senators that AI will significantly transform DoD’s capabilities and that it is critical the U.S. remain competitive with China and Russia in developing AI applications. Dr. Peter T. Highnam on DARPA’s achievements and future goals Dr. Peter T. Highnam, Deputy Director, Defense Advanced Research Projects Agency talked about DARPA’s significant role in the development of AI technologies that have produced game-changing capabilities for the Department of Defense and beyond. In his testimony, he mentions, “DARPA’s AI Next effort is simply a continuing part of its 166 historic investment in the exploration and advancement of AI technologies.” Dr. Highnam highlighted different waves of AI technologies. The first wave, which was nearly 70 years ago, emphasized handcrafted knowledge, and computer scientists constructed so-called expert systems that captured the rules that the system could then apply to situations of interest. However, handcrafting rules was costly and time-consuming. The second wave that brought in machine learning that applies statistical and probabilistic methods to large data sets to create generalized representations that can be applied to future samples. However, this required training deep learning (artificial) neural networks with a variety of classification and prediction tasks when adequate historical data. Therein lies the rub, however, as the task of collecting, labelling, and vetting data on which to train. Such a process is prohibitively costly and time-consuming too. He says, “DARPA envisions a future in which machines are more than just tools that execute human programmed rules or generalize from human-curated data sets. Rather, the machines DARPA envisions will function more as colleagues than as tools.” Towards this end, DARPA is focusing its investments on a “third wave” of AI technologies that brings forth machines that can reason in context. Incorporating these technologies in military systems that collaborate with warfighters will facilitate better decisions in complex, time-critical, battlefield environments; enable a shared understanding of massive, incomplete, and contradictory information; and empower unmanned systems to perform critical missions safely and with high degrees of autonomy. DARPA’s more than $2 billion “AI Next” campaign, announced in September 2018, includes providing robust foundations for second wave technologies, aggressively applying the second wave AI technologies into appropriate systems, and exploring and creating third wave AI science and technologies. DARPA’s third wave research efforts will forge new theories and methods that will make it possible for machines to adapt contextually to changing situations, advancing computers from tools to true collaborative partners. Furthermore, the agency will be fearless about exploring these new technologies and their capabilities – DARPA’s core function – pushing critical frontiers ahead of our nation’s adversaries. To know more about this in detail, read Dr. Peter T. Highnam’s complete statement. Michael Brown on (Defense Innovation Unit) DIU’s efforts in Artificial Intelligence Michael Brown, Director of the Defense Innovation Unit, started the talk by highlighting on the fact how China and Russia are investing heavily to become dominant in AI.  “By 2025, China will aim to achieve major breakthroughs in AI and increase its domestic market to reach $59.6 billion (RMB 400 billion) To achieve these targets, China’s National Development and Reform Commission (China’s industrial policy-making agency) funded the creation of a national AI laboratory, and Chinese local governments have pledged more than $7 billion in AI funding”, Brown said in his statement. He said that these Chinese firms are in a way leveraging U.S. talent by setting up research institutes in the state, investing in U.S. AI-related startups and firms, recruiting U.S.-based talent, and commercial and academic partnerships. Brown said that DIU will engage with DARPA and JAIC(Joint Artificial Intelligence Center) and also make its commercial knowledge and relationships with potential vendors available to any of the Services and Service Labs. DIU also anticipates that with its close partnership with the JAIC, DIU will be at the leading edge of the Department’s National Mission Initiatives (NMIs), proving that commercial technology can be applied to critical national security challenges via accelerated prototypes that lay the groundwork for future scaling through JAIC. “DIU looks to bring in key elements of AI development pursued by the commercial sector, which relies heavily on continuous feedback loops, vigorous experimentation using data, and iterative development, all to achieve the measurable outcome, mission impact”, Brown mentions. DIU’s AI portfolio team combines depth of commercial AI, machine learning, and data science experience from the commercial sector with military operators. However, they have specifically prioritized projects that address three major impact areas or use cases which employ AI technology, including: Computer vision The DIU is prototyping computer vision algorithms in humanitarian assistance and disaster recovery scenarios. “This use of AI holds the potential to automate post-disaster assessments and accelerate search and rescue efforts on a global scale”, Brown said in his statement. Large dataset analytics and predictions DIU is prototyping predictive maintenance applications for Air Force and Army platforms. For this DIU plans to partner with JAIC to scale this solution across multiple aircraft platforms, as well as ground vehicles beginning with DIU’s complementary predictive maintenance project focusing on the Army’s Bradley Fighting Vehicle. Brown says this is one of DIU’s highest priority projects for FY19 given its enormous potential for impact on readiness and reducing costs. Strategic reasoning DIU is prototyping an application from Project VOLTRON that leverages AI to reason about high-level strategic questions, map probabilistic chains of events, and develop alternative strategies. This will make DoD owned systems more resilient to cyber attacks and inform program offices of configuration errors faster and with fewer errors than humans. Know more about what more DIU plans in partnership with DARPA and JAIC, in detail, in Michael Brown’s complete testimony. Lieutenant General Jack Shanahan on making JAIC “AI-Ready” Lieutenant General Jack Shanahan, Director, Joint Artificial Intelligence Center, touches upon  how the JAIC is partnering with the Under Secretary of Defense (USD) Research & Engineering (R&E), the role of the Military Services, the Department’s initial focus areas for AI delivery, and how JAIC is supporting whole-of-government efforts in AI. “To derive maximum value from AI application throughout the Department, JAIC will operate across an end-to-end lifecycle of problem identification, prototyping, integration, scaling, transition, and sustainment. Emphasizing commerciality to the maximum extent practicable, JAIC will partner with the Services and other components across the Joint Force to systematically identify, prioritize, and select new AI mission initiatives”, Shanahan mentions in his testimony. The AI capability delivery efforts that will go through this lifecycle will fall into two categories including National Mission Initiatives (NMI) and Component Mission Initiatives (CMI). NMI is an operational or business reform joint challenge, typically identified from the National Defense Strategy’s key operational problems and requiring multi-service innovation, coordination, and the parallel introduction of new technology and new operating concepts. On the other hand, Component Mission Initiatives (CMI) is a component-level challenge that can be solved through AI. JAIC will work closely with individual components on CMIs to help identify, shape, and accelerate their Component-specific AI deployments through: funding support; usage of common foundational tools, libraries, cloud infrastructure; application of best practices; partnerships with industry and academia; and so on. The Component will be responsible for identifying and implementing the organizational structure required to accomplish its project in coordination and partnership with the JAIC. Following are some examples of early NMI’s by JAIC to deliver mission impact at speed, demonstrate the proof of concept for the JAIC operational model, enable rapid learning and iterative process refinement, and build their library of reusable tools while validating JAIC’s enterprise cloud architecture. Perception Improve the speed, completeness, and accuracy of Intelligence, Surveillance, Reconnaissance (ISR) Processing, Exploitation, and Dissemination (PED). Shanahan says Project Maven’s efforts are included here. Predictive Maintenance (PMx) Provide computational tools to decision-makers to help them better forecast, diagnose, and manage maintenance issues to increase availability, improve operational effectiveness, and ensure safety, at a reduced cost. Humanitarian Assistance/Disaster Relief (HA/DR) Reduce the time associated with search and discovery, resource allocation decisions, and executing rescue and relief operations to save lives and livelihood during disaster operations. Here, JAIC plans to apply lessons learned and reusable tools from Project Maven to field AI capabilities in support of federal responses to events such as wildfires and hurricanes—where DoD plays a supporting role. Cyber Sensemaking Detect and deter advanced adversarial cyber actors who infiltrate and operate within the DoD Information Network (DoDIN) to increase DoDIN security, safeguard sensitive information, and allow warfighters and engineers to focus on strategic analysis and response. Shanahan states, “Under the DoD CIO’s authorities and as delineated in the JAIC establishment memo, JAIC will coordinate all DoD AI-related projects above $15 million annually.” “It does mean that we will start to ensure, for example, that they begin to leverage common tools and libraries, manage data using best practices, reflect a common governance framework, adhere to rigorous testing and evaluation methodologies, share lessons learned, and comply with architectural principles and standards that enable scale”, he further added. To know more about this in detail, read Lieutenant General Jack Shanahan’s complete testimony. To know more about this news in detail, watch the entire hearing on 'Artificial Intelligence Initiatives within the Department of Defense' So, you want to learn artificial intelligence. Here’s how you do it. What can happen when artificial intelligence decides on your loan request Mozilla partners with Ubisoft to Clever-Commit its code, an artificial intelligence assisted assistant

The Google Walkout for Real Change group tweeted out their protest against the news of ‘Google confirming that it paid $135 million as exit packages to the two top execs accused of sexual assault, on Twitter, earlier this week. The group castigated the ‘multi-million dollar payouts’ and asked people to use the hashtag #GooglePayoutsForAll to demonstrate different and better ways this obscenely large amount of ‘hush money’ could have been used. https://twitter.com/GoogleWalkout/status/1105556617662214145 The news of Google paying its senior execs, namely, Amit Singhal (former Senior VP of Google search) and Andy Rubin (creator of Android) high exit packages was first highlighted in a report by the New York Times, last October. As per the report, Google paid $90 million to Rubin and $15 million to Singhal. A lawsuit filed by James Martin, an Alphabet shareholder, on Monday this week, further confirmed this news. The lawsuit states that this decision taken by directors of Alphabet caused significant financial harm to the company apart from deteriorating its reputation, goodwill, and market capitalization. Meredith Whittaker, one of the early organizers of the Google Walkout in November last month tweeted, “$135 million could fix Flint's water crisis and still have $80 million left.” Vicki Tardif, another Googler summed up the sentiments in her tweet, “$135M is 1.35 times what Google.org  gave out in grants in 2016.” An ACLU researcher pointed out that $135M could have in addition to feeding the hungry, housing the homeless and pay off some student loans, It could also support local journalism killed by online ads. The public support to the call for protest using the hashtag #GooglePayoutsForAll has been awe-inspiring. Some shared their stories of injustice in cases of sexual assault, some condemned Google for its handling of sexual misconduct, while others put the amount of money Google wasted on these execs into a larger perspective. Better ways Google could have used $135 million it wasted on execs payouts, according to Twitter Invest in people to reduce structural inequities in the company $135M could have been paid to the actual victims who faced harassment and sexual assault. https://twitter.com/xzzzxxzx/status/1105681517584572416 Google could have used the money to fix the wage and level gap for women of color within the company. https://twitter.com/sparker2/status/1105511306465992705 $135 million could be used to adjust the 16% median pay gap of the 1240 women working in Google’s UK offices https://twitter.com/crschmidt/status/1105645484104998913 $135M could have been used by Google for TVC benefits. It could also be used to provide rigorous training to the Google employees on what impact misinformation within the company can have on women and other marginalized groups.   https://twitter.com/EricaAmerica/status/1105546835526107136 For $135M, Google could have paid the 114 creators featured in its annual "YouTube Rewind" who are otherwise unpaid for their time and participation. https://twitter.com/crschmidt/status/1105641872033230848 Improve communities by supporting social causes Google could have paid $135M to RAINN, a largest American nonprofit anti-sexual assault organization, covering its expenses for the next 18 years. https://twitter.com/GoogleWalkout/status/1105450565193121792 For funding 1800 school psychologists for 1 year in public schools https://twitter.com/markfickett/status/1105640930936324097 To build real, affordable housing solutions in collaboration with London Breed, SFGOV, and other Bay Area officials https://twitter.com/jillianpuente/status/1105922474930245636 $135M could provide insulin for nearly 10,000 people with Type 1 diabetes in the US https://twitter.com/GoogleWalkout/status/1105585078590210051 To pay for the first year for 1,000 people with stage IV breast cancer https://twitter.com/GoogleWalkout/status/1105845951938347008 Be a responsible corporate citizen To fund approximately 5300 low-cost electric vehicles for Google staff, and saving around 25300 metric tons of carbon dioxide from vehicle emissions per year. https://twitter.com/crschmidt/status/1105698893361233926 Providing free Google Fiber internet to 225,000 homes for a year https://twitter.com/markfickett/status/1105641215389773825 To give $5/hr raise to 12,980 service workers at Silicon Valley tech campuses https://twitter.com/LAuerhahn/status/1105487572069801985 $135M could have been used for the construction of affordable homes, protecting 1,100 low-income families in San Jose from coming rent hikes of Google’s planned mega-campus. https://twitter.com/JRBinSV/status/1105478979543154688 #GooglePayoutsForAll: Another initiative to promote awareness of structural inequities in tech   The core idea behind launching #GooglePayoutsForAll on Twitter by the Google walkout group was to promote awareness among people regarding the real issues within the company. It urged people to discuss how Google is failing at maintaining the ‘open culture’ that it promises to the outside world. It also highlights how mottos such as “Don’t be Evil” and “Do the right thing” that Google stood by only make for pretty wall decor and there’s still a long way to go to see those ideals in action. The group gained its name when more than 20,000 Google employees along with vendors, and contractors, temps, organized Google “walkout for real change” and walked out of their offices in November 2018. The walkout was a protest against the hushed and unfair handling of sexual misconduct within Google. Ever since then, Googlers have been consistently taking initiatives to bring more transparency, accountability, and fairness within the company. For instance, the team launched an industry-wide awareness campaign to fight against forced arbitration in January, where they shared information about arbitration on their Twitter and Instagram accounts throughout the day. The campaign was a success as Google finally ended its forced arbitration policy which goes into effect this month for all the employees (including contractors, temps, vendors) and for all kinds of discrimination. Also, House and Senate members in the US have proposed a bipartisan bill to prohibit companies from using forced arbitration clauses, last month.    Although many found the #GooglePayoutsForAll idea praiseworthy, some believe this initiative doesn’t put any real pressure on Google to bring about a real change within the company. https://twitter.com/Jeffanie16/status/1105541489722081290 https://twitter.com/Jeffanie16/status/1105546783063752709 https://twitter.com/Jeffanie16/status/1105547341862457344 Now, we don’t necessarily disagree with this opinion, however, the initiative can't be completely disregarded as it managed to make people who’d otherwise hesitate to open up talk extensively regarding the real issues within the company. As Liz Fong-Jones puts it, “Strikes and walkouts are more sustainable long-term than letting Google drive each organizer out one by one. But yes, people *are* taking action in addition to speaking up. And speaking up is a bold step in companies where workers haven't spoken up before”. The Google Walkout group have not yet announced what they intend to do next following this digital protest. However, the group has been organizing meetups such as the one earlier this month on March 6th where it invited the tech contract workers for discussion about building solidarity to make work better for everyone. We are only seeing the beginning of a powerful worker movement take shape in Silicon Valley. Recode Decode #GoogleWalkout interview shows why data and evidence don’t always lead to right decisions in even the world’s most data-driven company Liz Fong Jones, prominent ex-Googler shares her experience at Google and ‘grave concerns’ for the company Google’s pay equity analysis finds men, not women, are underpaid; critics call out design flaws in the analysis

According to a complaint filed in a lawsuit yesterday, Google paid $135 million in total as exit packages to top two senior execs, namely Andy Rubin (creator of Android) and Amit Singhal (former senior VP of Google search) after they were accused of sexual misconduct in the company. The lawsuit was filed by an Alphabet shareholder, James Martin, in the Santa Clara, California Court. Google also confirmed paying the exit packages to senior execs to The Verge, yesterday. Speaking of the lawsuit, the complaint is against certain directors and officers of Alphabet, Google’s parent company, for their active and direct participation in “multi-year scheme” to hide sexual harassment and discrimination at Alphabet. It also states that the misconduct by these directors has caused severe financial and reputational damage to Alphabet. The exit packages for Rubin and Singhal were approved by the Leadership Development and Compensation Committee (LLDC). The news of Google paying high exit packages to its top execs first came to light last October, after the New York Times released a report on Google, stating that the firm paid $90 million to Rubin and $15 million to Singhal. Rubin had previously also received an offer for a $150 million stock grant, which he then further use to negotiate the $90 million in severance pay, even though he should have been fired for cause without any pay, states the lawsuit. To protest against the handling of sexual misconduct within Google, more than 20,000 Google employees along with vendors, and contractors, temps, organized Google “walkout for real change” and walked out of their offices in November 2018. Googlers also launched an industry-wide awareness campaign to fight against forced arbitration in January, where they shared information about arbitration on their Twitter and Instagram accounts throughout the day.   Last year in November, Google ended its forced arbitration ( a move that was soon followed by Facebook) for its employees (excluding temps, vendors, etc) and only in the case of sexual harassment. This led to contractors writing an open letter on Medium to Sundar Pichai, CEO, Google, in December, demanding him to address their demands of better conditions and equal benefits for contractors. In response to the Google walkout and the growing public pressure, Google finally decided to end its forced arbitration policy for all employees (including contractors) and for all kinds of discrimination within Google, last month. The changes will go into effect for all the Google employees starting March 21st, 2019. Yesterday, the Google walkout for real change group tweeted condemning the multi-million dollar payouts and has asked people to use the hashtag #Googlepayoutsforall to highlight other better ways that money could have been used. https://twitter.com/GoogleWalkout/status/1105450565193121792 “The conduct of Rubin and other executives was disgusting, illegal, immoral, degrading to women and contrary to every principle that Google claims it abides by”, reads the lawsuit. James Martin also filed a lawsuit against Alphabet’s board members, Larry Page, Sergey Brin, and Eric Schmidt earlier this year in January for covering up the sexual harassment allegations against the former top execs at Google. Martin had sued Alphabet for breaching its fiduciary duty to shareholders, unjust enrichment, abuse of power, and corporate waste. “The directors’ wrongful conduct allowed illegal conduct to proliferate and continue. As such, members of the Alphabet’s board were knowing direct enables of sexual harassment and discrimination”, reads the lawsuit. It also states that the board members not only violated the California and federal law but it also violated the ethical standards and guidelines set by Alphabet. Public reaction to the news is largely negative with people condemning Google’s handling of sexual misconduct: https://twitter.com/awesome/status/1105295877487263744 https://twitter.com/justkelly_ok/status/1105456081663225856 https://twitter.com/justkelly_ok/status/1105457965790707713 https://twitter.com/conradwt/status/1105386882135875584 https://twitter.com/mer__edith/status/1105464808831361025 For more information, check out the official lawsuit here. Recode Decode #GoogleWalkout interview shows why data and evidence don’t always lead to right decisions in even the world’s most data-driven company Liz Fong Jones, prominent ex-Googler shares her experience at Google and ‘grave concerns’ for the company Google’s pay equity analysis finds men, not women, are underpaid; critics call out design flaws in the analysis

The two-days long TensorFlow Dev Summit 2019 just got over, leaving in its wake major updates being made to the TensorFlow ecosystem.  The major announcement included the release of the first alpha version of most coveted release TensorFlow 2.0. Also announced were, TensorFlow Lite 1.0, TensorFlow Federated, TensorFlow Privacy and more. TensorFlow Federated In a medium blog post, Alex Ingerman (Product Manager) and Krzys Ostrowski (Research Scientist) introduced the TensorFlow Federated framework on the first day. This open source framework is useful for experimenting with machine learning and other computations on decentralized data. As the name suggests, this framework uses Federated Learning, a learning approach introduced by Google in 2017. This technique enables ML models to collaboratively learn a shared prediction model while keeping all the training data on the device. Thus eliminating machine learning from the need to store the data in the cloud. The authors note that TFF is based on their experiences with developing federated learning technology at Google. TFF uses the Federated Learning API to express an ML model architecture, and then train it across data provided by multiple developers, while keeping each developer’s data separate and local. It also uses the Federated Core (FC) API, a set of lower-level primitives, which enables the expression of a broad range of computations over a decentralized dataset. The authors conclude, “With TFF, we are excited to put a flexible, open framework for locally simulating decentralized computations into the hands of all TensorFlow users. You can try out TFF in your browser, with just a few clicks, by walking through the tutorials.” TensorFlow 2.0.0- alpha0 The event also the release of the first alpha version of the TensorFlow 2.0 framework which came with fewer APIs. First introduced last year in August by Martin Wicke, engineer at Google, TensorFlow 2.0, is expected to come with: Easy model building with Keras and eager execution. Robust model deployment in production on any platform. Powerful experimentation for research. API simplification by reducing duplication removing deprecated endpoints. The first teaser,  TensorFlow 2.0.0- alpha0 version comes with the following changes: API clean-up included removing tf.app, tf.flags, and tf.logging in favor of absl-py. No more global variables with helper methods like tf.global_variables_initializer and tf.get_global_step. Functions, not sessions (tf.Session and session.run -> tf.function). Added support for TensorFlow Lite in TensorFlow 2.0. tf.contrib has been deprecated, and functionality has been either migrated to the core TensorFlow API, to tensorflow/addons, or removed entirely. Checkpoint breakage for RNNs and for Optimizers. Minor bug fixes have also been made to the Keras and Python API and tf.estimator. Read the full list of bug fixes in the changelog. TensorFlow Lite 1.0 The TF-Lite framework is basically designed to aid developers in deploying machine learning and artificial intelligence models on mobile and IoT devices. Lite was first introduced at the I/O developer conference in May 2017 and in developer preview later that year. At the TensorFlow Dev Summit, the team announced a new version of this framework, the TensorFlow Lite 1.0. According to a post by VentureBeat, improvements include selective registration and quantization during and after training for faster, smaller models. The team behind TF-Lite 1.0 says that quantization has helped them achieve up to 4 times compression of some models. TensorFlow Privacy Another interesting library released at the TensorFlow dev summit was TensorFlow Privacy. This Python-based open source library aids developers to train their machine-learning models with strong privacy guarantees. To achieve this, it takes inspiration from the principles of differential privacy. This technique offers strong mathematical guarantees that models do not learn or remember the details about any specific user when training the user data. TensorFlow Privacy includes implementations of TensorFlow optimizers for training machine learning models with differential privacy. For more information, you can go through the technical whitepaper describing its privacy mechanisms in more detail. The creators also note that “no expertise in privacy or its underlying mathematics should be required for using TensorFlow Privacy. Those using standard TensorFlow mechanisms should not have to change their model architectures, training procedures, or processes.” TensorFlow Replicator TF Replicator also released at the TensorFlow Dev Summit, is a software library that helps researchers deploy their TensorFlow models on GPUs and Cloud TPUs. To do this, the creators assure that developers would require minimal effort and need not have previous experience with distributed systems. For multi-GPU computation, TF-Replicator relies on an “in-graph replication” pattern, where the computation for each device is replicated in the same TensorFlow graph. When TF-Replicator builds an in-graph replicated computation, it first builds the computation for each device independently and leaves placeholders where cross-device computation has been specified by the user. Once the sub-graphs for all devices have been built, TF-Replicator connects them by replacing the placeholders with actual cross-device computation. For a more comprehensive description, you can go through the research paper. These were the top announcements made at the TensorFlow Dev Summit 2019. You can go through the Keynote and other videos of the announcements and tutorials on this YouTube playlist. TensorFlow 2.0 to be released soon with eager execution, removal of redundant APIs, tffunction and more. TensorFlow 2.0 is coming. Here’s what we can expect. Google introduces and open-sources Lingvo, a scalable TensorFlow framework for Sequence-to-Sequence Modeling

In this article, we will provide 3 tips for making an interactive conversational application using current chat and voice examples. This is an excerpt from the book Voicebot and Chatbot Design written by Rachel Batish. In this book, the author shares her insights into cutting-edge voice-bot and chatbot technologies Help your users ask the right questions Although this sounds obvious, it is actually crucial to the success of your chatbot or voice-bot. I learned this when I initially set up my Amazon Echo device at home. Using a complementary mobile app, I was directed to ask Alexa specific questions, to which she had good answers to, such as “Alexa, what is the time?” or “Alexa, what is the weather today?” I immediately received correct answers and therefore wasn’t discouraged by a default response saying, “Sorry, I don’t have an answer to that question.” By providing the user with successful experience, we are encouraging them to trust the system and to understand that, although it has its limitations, it is really good in some specific details. Obviously, this isn’t enough because as time passes, Alexa (and Google) continues to evolve and continues to expand its support and capabilities, both internally and by leveraging third parties. To solve this discovery problem, some solutions, like Amazon Alexa and Google Home, send a weekly newsletter with the highlights of their latest capabilities. In the email below, Amazon Alexa is providing a list of questions that I should ask Alexa in my next interaction with it, exposing me to new functionalities like donation. From the Amazon Alexa weekly emails “What’s new with Alexa?” On the Google Home/Assistant, Google has also chosen topics that it recommends its users to interact with. Here, as well, the end user is exposed to new offerings/capabilities/knowledge bases, that may give them the trust needed to ask similar questions on other topics. From the Google Home newsletter Other chat and voice providers can also take advantage of this email communication idea to encourage their users to further interact with their chatbots or voice-bots and expose them to new capabilities. The simplest way of encouraging usage is by adding a dynamic ‘welcoming’ message to the chat voice applications, that includes new features that are enabled. Capital One, for example, updates this information every now and then, exposing its users to new functionalities. On Alexa, it sounds like this: “Welcome to Capital One. You can ask me for things like account balance and recent transactions.” Another way to do this – especially if you are reaching out to a random group of people – is to initiate discovery during the interaction with the user (I call this contextual discovery). For example, a banking chatbot offers information on account balances. Imagine that the user asks, “What’s my account balance?” The system gives its response: “Your checking account balance is $5,000 USD.” The bank has recently activated the option to transfer money between accounts. To expose this information to its users, it leverages the bot to prompt a rational suggestion to the user and say, “Did you know you can now transfer money between accounts? Would you like me to transfer $1,000 to your savings account?” As you can see, the discovery process was done in context with the user’s actions. Not only does the user know that he/she can now transfer money between two accounts, but they can also experience it immediately, within the relevant context. To sum up tip #1, by finding the direct path to initial success, your users will be encouraged to further explore and discover your automated solutions and will not fall back to other channels. The challenge is, of course, to continuously expose users to new functionalities, made available on your chatbots and voice-bots, preferably in a contextual manner. Give your bot a ‘personality’, but don’t pretend it’s a human Your bot, just like any digital solution you provide today, should have a personality that makes sense for your brand. It can be visual, but it can also be enabled over voice. Whether it is a character you use for your brand or something created for your bot, personality is more than just the bot’s icon. It’s the language that it ‘speaks’, the type of interaction that it has and the environment it creates. In any case, don’t try to pretend that your bot is a human talking with your clients. People tend to ask the bot questions like “are you a bot?” and sometimes even try to make it fail by asking questions that are not related to the conversation (like asking how much 30*4,000 is or what the bot thinks of *a specific event*). Let your users know that it’s a bot that they are talking to and that it’s here to help. This way, the user has no incentive to intentionally trip up the bot. ICS.ai have created many custom bots for some of the leading UK public sector organisations like county councils, local governments and healthcare trusts. Their conversational AI chat bots are custom designed by name, appearance and language according to customer needs. Chatbot examples Below are a few examples of chatbots with matching personalities. Expand your vocabulary with a word a day (Wordsworth) The Wordsworth bot has a personality of an owl (something clever), which fits very well with the purpose of the bot: to enrich the user’s vocabulary. However, we can see that this bot has more than just an owl as its ‘presenter’, pay attention to the language and word games and even the joke at the end. Jokes are a great way to deliver personality. From these two screenshots only, we can easily capture a specific image of this bot, what it represents and what it’s here to do. DIY-Crafts-Handmade FB Messenger bot The DIY-Crafts-Handmade bot has a different personality, which signals something light and fun. The language used is much more conversational (and less didactic) and there’s a lot of usage of icons and emojis. It’s clear that this bot was created for girls/women and offers the end user a close ‘friend’ to help them maximize the time they spend at home with the kids or just start some DIY projects. Voicebot examples One of the limitations around today’s voice-enabled devices is the voice itself. Whereas Google and Siri do offer a couple of voices to choose from, Alexa is limited to only one voice and it’s very difficult to create that personality that we are looking for. While this problem probably will be solved in the future, as technology improves, I find insurance company GEICO’s creativity around that very inspiring. In its effort to keep Gecko’s unique voice and personality, GEICO has incorporated multiple MP3 files with a recording of Gecko’s personalized voice. https://www.youtube.com/watch?v=11qo9a1lgBE GEICO has been investing for years in Gecko’s personalization. Gecko is very familiar from TV and radio advertisements, so when a customer activates the Alexa app or Google Action, they know they are in the right place. To make this successful, GEICO incorporated Gecko’s voice into various (non-dynamic) messages and greetings. It also handled the transition back to the device’s generic voice very nicely; after Gecko has greeted the user and provided information on what they can do, it hands it back to Alexa with every question from the user by saying, “My friend here can help you with that.” This is a great example of a cross-channel brand personality that comes to life also on automated solutions such as chatbots and voice-bots. Build an omnichannel solution – find your tool Think less on the design side and more on the strategic side, remember that new devices are not replacing old devices; they are only adding to the big basket of channels that you must support. Users today are looking for different services anywhere and anytime. Providing a similar level of service on all the different channels is not an easy task, but it will play a big part in the success of your application. There are different reasons for this. For instance, you might see a spike in requests coming from home devices such as Amazon Echo and Google Home during the early morning and late at night. However, during the day you will receive more activities from FB Messenger or your intelligent assistant. Different age groups also consume products from different channels and, of course, geography impacts as well. Providing cross-channel/omnichannel support doesn’t mean providing different experiences or capabilities. However, it does mean that you need to make that extra effort to identify the added value of each solution, in order to provide a premium, or at least the most advanced, experience on each channel. Building an omnichannel solution for voice and chat Obviously, there are differences between a chatbot and a voice-bot interaction; we talk differently to how we write and we can express ourselves with emojis while transferring our feelings with voice is still impossible. There are even differences between various voice-enabled devices, like Amazon Alexa and Google Assistant/Home and, of course, Apple’s HomePod. There are technical differences but also behavioral ones. The HomePod offers a set of limited use cases that businesses can connect with, whereas Amazon Alexa and Google Home let us create our own use cases freely. In fact, there are differences between various Amazon Echo devices, like the Alexa Show that offers a complimentary screen and the Echo Dot that lacks in screen and sound in comparison. There are some developer tools today that offer multi-channel integration to some devices and channels. They are highly recommended from a short and long-term perspective. Those platforms let bot designers and bot builders focus on the business logic and structure of their bots, while all the integration efforts are taken care of automatically. Some of those platforms focus on chat and some of them on voice. A few tools offer a bridge between all the automated channels or devices. Among those platforms, you can find Conversation.one (disclaimer: I’m one of the founders), Dexter and Jovo. With all that in mind, it is clear that developing a good conversational application is not an easy task. Developers must prove profound knowledge of machine learning, voice recognition, and natural language processing. In addition to that, it requires highly sophisticated and rare skills, that are extremely dynamic and flexible. In such a high-risk environment, where today’s top trends can skyrocket in days or simply be crushed in just a few months, any initial investment can be dicey. To know more trips and tricks to make a successful chatbot or voice-bot, read the book Voicebot and Chatbot Design by Rachel Batish. Creating a chatbot to assist in network operations [Tutorial] Building your first chatbot using Chatfuel with no code [Tutorial] Conversational AI in 2018: An arms race of new products, acquisitions, and more

In theory, stealing cryptocurrency should be impossible. But a mystery has emerged that seems to throw all that into question and even suggests a bigger, much stranger conspiracy. Gerald Cotten, the founder of cryptocurrency exchange QadrigaCX, died in December in India. He was believed to have left $136 million USD worth of crypto-cash in 'cold wallets' on his own laptop, to which only he had access. However, investigators from EY, who have been working on closing QuadrigaCX following Cotten's death, were surprised to find that the wallets were empty. In fact, it's believed crypto-cash had disappeared from them months before Cotten died. A cryptocurrency mystery now involving the FBI The only lead in this mystery is the fact that the EY investigators have found other user accounts that appear to be linked to Gerald Cotten. There's a chance that Cotten used these to trade on his own exchange, but the nature of these exchanges remain a little unclear. To add to the intrigue, Fortune reported yesterday that the FBI are working with Canada's Mounted Police Force to investigate the missing money. This information came from Jesse Powell, CEO of another cryptocurrency company called Kraken. Powell told Fortune that both the FBI and the Mounted Police have been in touch with him about the mystery surrounding QuadrigaCX. Powell has offered a reward of $100,000 to anyone that can locate the missing cryptocurrency funds. So what actually happened to Gerald Cotten and his crypto-cash? The story has many layers of complexity. There are rumors that Cotten faked his own death. For example, Cotten filed a will just 12 days before his death, leaving a significant amount of wealth and assets to his wife. And while sources from the hospital in India where Cotten is believed to have died say he died of cardiac arrest, as Fortune explains, "Cotten’s body was handled by hotel staff after an embalmer refused to receive it" - something which is, at the very least, strange. It should be noted that there is certainly no clear evidence that Cotten faked his own death - only missing pieces that encourage such rumors. A further subplot - that might or night not be useful in cracking this case - emerged late last week when Canada's Globe and Mail reported that QuadrigaCX's co-founder has a history of identity theft and using digital currencies to launder money. Where could the money be? There are, as you might expect, no shortage of theories about where the cash could be. A few days ago, it was suggested that it might be possible to locate Cotten's Ethereum funds - a blog post by James Edwards, who is the editor of cryptocurrency blog zerononcense claimed that Ethereum linked to QuadrigaCX can be found in Bitfinex, Poloniex, and Jesse Powell's Kraken. "It appears that a significant amount of Ethereum (600,000+ ETH) was transferred to these exchanges as a means of ‘storage’ during the years that QuadrigaCX was in operation and offering Ethereum on their exchange," Edwards writes. Edwards is keen for his findings to be the starting point for a clearer line of inquiry, free from speculation and conspiracy. He wrote that he hoped that it would be "a helpful addition to the QuadrigaCX narrative, rather than a conspiratorial piece that speculates on whether the exchange or its owners have been honest."

It’s not always easy to know what to learn next if you’re a programmer. Industry shifts can be subtle but they can sometimes be dramatic, making it incredibly important to stay on top of what’s happening both in your field and beyond. No one person can make that decision for you. All the thought leadership and mentorship in the world isn’t going to be able to tell you what’s right for you when it comes to your career. But this list of videos, released last month, might give you a helping hand as to where to go next when it comes to your learning… New data science and artificial intelligence video courses for March Apache Spark is carving out a big presence as the go-to software for big data. Two videos from February focus on Spark - Distributed Deep Learning with Apache Spark and Apache Spark in 7 Days. If you’re new to Spark and want a crash course on the tool, then clearly, our video aims to get you up and running quickly. However, Distributed Deep Learning with Apache Spark offers a deeper exploration that shows you how to develop end to end deep learning pipelines that can leverage the full potential of cutting edge deep learning techniques. While we’re on the subject of machine learning, other choice video courses for March include TensorFlow 2.0 New Features (we’ve been eagerly awaiting it and it finally looks like we can see what it will be like), Hands On Machine Learning with JavaScript (yes, you can now do machine learning in the browser), and a handful of interesting videos on artificial intelligence and finance: AI for Finance Machine Learning for Algorithmic Trading Bots with Python Hands on Python for Finance Elsewhere, a number of data visualization video courses prove that communicating and presenting data remains an urgent challenge for those in the data space. Tableau remains one of the definitive tools - you can learn the latest version with Tableau 2019.1 for Data Scientists and Data Visualization Recipes with Python and Matplotlib 3.   New app and web development video courses for March 2019 There are a wealth of video courses for web and app developers to choose from this month. True, Hands-on Machine Learning for JavaScript is well worth a look, but moving past the machine learning hype, there are a number of video courses that take a practical look at popular tools and new approaches to app and web development. Angular’s death has been greatly exaggerated - it remains a pillar of the JavaScript world. While the project’s versioning has arguably been lacking some clarity, if you want to get up to speed with where the framework is today, try Angular 7: A Practical Guide. It’s a video that does exactly what it says on the proverbial tin - it shows off Angular 7 and demonstrates how to start using it in web projects. We’ve also been seeing some uptake of Angular by ASP.NET developers, as it offers a nice complement to the Microsoft framework on the front end side. Our latest video on the combination, Hands-on Web Development with ASP.NET Core and Angular, is another practical look at an effective and increasingly popular approach to full-stack development. Other picks for March include Building Mobile Apps with Ionic 4, a video that brings you right up to date with the recent update that launched in January (interestingly, the project is now backed by web components, not Angular), and a couple of Redux videos - Mastering Redux and Redux Recipes. Redux is still relatively new. Essentially, it’s a JavaScript library that helps you manage application state - because it can be used with a range of different frameworks and libraries, including both Angular and React, it’s likely to go from strength to strength in 2019. Infrastructure, admin and security video courses for March 2019 Node.js is becoming an important library for infrastructure and DevOps engineers. As we move to a cloud native world, it’s a great tool for developing lightweight and modular services. That’s why we’re picking Learn Serverless App Development with Node.js and Azure Functions as one of our top videos for this month. Azure has been growing at a rapid rate over the last 12 months, and while it’s still some way behind AWS, Microsoft’s focus on developer experience is making Azure an increasingly popular platform with developers. For Node developers, this video is a great place to begin - it’s also useful for anyone who simply wants to find out what serverless development actually feels like. Read next: Serverless computing wars: AWS Lambda vs. Azure Functions A partner to this, for anyone beginning Node, is the new Node.js Design Patterns video. In particular, if Node.js is an important tool in your architecture, following design patterns is a robust method of ensuring reliability and resilience. Elsewhere, we have Modern DevOps in Practice, cutting through the consultancy-speak to give you useful and applicable guidance on how to use DevOps thinking in your workflows and processes, and DevOps with Azure, another video that again demonstrates just how impressive Azure is. For those not Azure-inclined, there’s AWS Certified Developer Associate - A Practical Guide, a video that takes you through everything you need to know to pass the AWS Developer Associate exam. There’s also a completely cloud-agnostic video course in the form of Creating a Continuous Deployment Pipeline for Cloud Platforms that’s essential for infrastructure and operations engineers getting to grips with cloud native development.     Learn a new programming language with these new video courses for March Finally, there are a number of new video courses that can help you get to grips with a new programming language. So, perfect if you’ve been putting off your new year’s resolution to learn a new language… Java 11 in 7 Days is a new video that brings you bang up to date with everything in the latest version of Java, while Hands-on Functional Programming with Java will help you rethink and reevaluate the way you use Java. Together, the two videos are a great way for Java developers to kick start their learning and update their skill set.  

In a paper published on February 4, Google engineers drafted out plans to forward federated learning at a scale. It showcases the high-level plans, challenges, solutions, and applications. Federated learning was first introduced in 2017 by Google. The idea is to use data from a number of computing devices like smartphones instead of a centralized data source. Federated learning can help with privacy Federated learning can be beneficial as it addresses the privacy concern. Android phones are used for the system where the data is only used but never uploaded to any server. A deep neural network is trained by using TensorFlow on the data stored in the Android phone. The Federated averaging algorithm by Brendan McMahan uses a similar approach as synchronous training. The weights of the neural network are combined in the cloud using Federated Averaging. This creates a global model which is then pushed back to the phones as results/desirable actions. To enhance privacy approaches like differential privacy and Secure aggregation are taken. The paper addresses challenges like time zone differences, connectivity issues, interrupted execution etc,. Their work is mature enough to deploy the system in production for tens of millions of devices. They are working towards supporting billions of devices now. The training protocol The system involves devices and the Federated Learning server communicating availability and the server selecting devices to run a task. A subset of the available devices are selected for a task. The Federated Learning server instructs the devices what computing task to run with a plan. A plan would consist a TensorFlow graph and instructions to execute it. There are three phases for the training to take place: Selection of the devices that meet eligibility criteria Configuring the server with simple or Secure Aggregation Reporting from the devices where reaching a certain number would get the training round started Source: Towards Federated Learning at Scale: System Design The devices are supposed to maintain a repository of the collected data and the applications are responsible to provide data to the Federated Learning runtime as an example store. The Federated Learning server is designed to operate on orders of many magnitudes. Each round can mean updates from devices in the range of KBs to tens of MBs coming going the server. Data collection To avoid harming the phone’s battery life and performance, various analytics are collected in the cloud. The logs don’t contain any personally identifiable information. Secure aggregation Secure aggregation uses encryption to make individual device updates uninspectable. They plant to use it for protection against threats in data centers. Secure aggregation would ensure data encryption even when it is in-memory. Challenges of federated learning Compared to a centralized dataset, federated learning poses a number of challenges. The training data is not inspectable, tooling is required to work with proxy data. Models cannot be run interactively and must be compiled to be deployed in the Federated Learning server. Model resource consumption and runtime compatibility also come into the picture when working with many devices in real-time. Applications of Federated Learning It is best for cases where the data on devices is more relevant than data on servers. Ranking items for better navigation, suggestions for on-device keyboard, and next word prediction. This has already been implemented on Google pixel and Gboard. Future work is to eliminate bias caused be restrictions in device selection, algorithms to support better parallelism (more devices in one round), avoiding retraining already trained tasks on devices, and compression to save bandwidth. Federated computation, not federated learning The authors do no mention machine learning explicitly anywhere in the paper. They believe that the applications of such a model are not limited to machine learning. Federated Computation is the term they want to use for this concept. Federated computation and edge computing Federated learning and edge computing are very similar, there are but subtle differences in the purpose of these two. Federated learning is used to solve problems with specific tasks assigned to endpoint smartphones. Edge computing is for predefined tasks to be processed at end nodes, for example, IoT cameras. Federated learning decentralizes the data used while edge computing decentralizes the task computation to various devices. For more details on the architecture and its working, you can check out the research paper. Technical and hidden debts in machine learning – Google engineers’ give their perspective Researchers introduce a machine learning model where the learning cannot be proved What if AIs could collaborate using human-like values? DeepMind researchers propose a Hanabi platform.

Kara Swisher, Recode co-founder, interviewed Jack Dorsey, Twitter CEO, yesterday over Twitter. The interview ( or ‘Twitterview’)  was conducted in tweets using the hashtag #KaraJack. It started at 5 pm ET and lasted for around 90-minutes. Let’s have a look at the top highlights from the interview. https://twitter.com/karaswisher/status/1095440667373899776 On Fixing what is broke on Social Media and Physical safety Swisher asked Dorsey why he isn’t moving faster in his efforts to fix the disaster that has been caused so far on social media. To this Dorsey replied that Twitter was trying to do “too much” in the past but that they have become better at prioritizing now. The number one focus for them now is a person’s “physical safety” i.e. the offline ramifications for Twitter users off the platform. “What people do offline with what they see online”, says Dorsey. Some examples of ‘offline ramifications’ being “doxxing” (harassment technique that reveals a person’s personal information on the internet) and coordinated harassment campaigns. Dorsey further added that replies, searches, trends, mentions on Twitter are where most of the abuse happens and are the shared spaces people take advantage of. “We need to put our physical safety above all else. We don’t have all the answers just yet. But that’s the focus. I think it clarifies a lot of the work we need to do. Not all of it of course”, said Dorsey. On Tech responsibility and improving the health of digital conversation on Twitter When Swisher asked Dorsey what grading would he give to Silicon Valley and himself for embodying tech responsibility, he replied with “C” for himself. He said that Twitter has made progress but it’s scattered and ‘not felt enough’. He did not comment on what he thought of Silicon Valley’s work in this area. Swisher further highlighted that the goal of improving Twitter conversations have only remained empty talk so far. She asked Dorsey if Twitter has made any actual progress in the last 18-24 months when it comes to addressing the issues regarding the “health of conversation” (which eventually plays into safety). Dorsey said these issues are the most important thing right now that they need to fix and it’s a failure on Twitter’s part to ‘put the burden on victims’. He did not share a specific example of improvements made to the platform to further this goal. Swisher then questioned him on how he intends on fixing the issue, Dorsey mentioned that: Twitter intends to be more proactive when it comes to enforcing healthy conversations so that reporting/blocking becomes the last resort. He mentioned that Twitter takes actions against all offenders who go against its policies but that the system works reactively to someone who reports it. “If they don’t report, we don’t see it. Doesn’t scale. Hence the need to focus on proactive”, said Dorsey. Since Twitter is constantly evolving its policies to address the ‘current issues’, it's rooting these in fundamental human rights (UN) and is making physical safety the top priority alongside privacy. On lack of diversity https://twitter.com/jack/status/1095459084785004544 Swisher questioned Dorsey on his negligence towards addressing the issues. “I think it is because many of the people who made Twitter never ever felt unsafe,” adds Swisher. Dorsey admits that the “lack of diversity” didn’t help with the empathy of what people (especially women) experience on Twitter every day. He further adds that Twitter should be reflective of the people that it’s trying to serve, which is why they established a trust and safety council to get feedback. Swisher then asks him to provide three concrete examples of what Twitter has done to fix this. Dorsey mentioned that Twitter has: evolved its policies ( eg; misgendering policy). prioritized proactive enforcement by using machine learning to downrank bad actors, meaning, they'll look at the probability of abuse from any one account. This is because if someone else is abusing one account then they’re probably doing the same on other accounts. Given more user control in a product, such as muting of accounts with no profile picture, etc. More focus on coordinated behavior/gaming. On Dorsey’s dual CEO role Swisher asked him why he insists on being the CEO of two publicly traded companies (Twitter and Square Inc.) that both require maximum effort at the same time. Dorsey said that his main focus is on building leadership in both and that it’s not his ambition to be CEO of multiple companies “just for the sake of that”. She further questioned him if he has any plans in mind to hire someone as his “number 2”. Dorsey said it’s better to spread that kind of responsibility across several people as it reduces dependencies and the company gets more options for future leadership. “I’m doing everything I can to help both. Effort doesn’t come down to one person. It’s a team”, he said. On Twitter breaks, Donald Trump and Elon Musk When initially asked about what Dorsey feels about people not feeling good after being for a while on Twitter, he said he feels “terrible” and that it's depressing. https://twitter.com/jack/status/1095457041844334593 “We made something with one intent. The world showed us how it wanted to use it. A lot has been great. A lot has been unexpected. A lot has been negative. We weren’t fast enough to observe, learn, and improve”, said Dorsey. He further added that he does not feel good about how Twitter tends to incentivize outrage, fast takes, short term thinking, echo chambers, and fragmented conversations. Swisher then questioned Dorsey on whether Twitter has ever intended on suspending Donald Trump and if Twitter’s business/engagement would suffer when Trump is no longer the president. Dorsey replied that Twitter is independent of any account or person and that although the number of politics conversations has increased on Twitter, that’s just one experience. He further added that Twitter is ready for 2020 elections and that it has partnered up with government agencies to improve communication around threats. https://twitter.com/jack/status/1095462610462433280 Moreover, on being asked about the most exciting influential on Twitter, Dorsey replied with Elon Musk. He said he likes how Elon is focused on solving existential problems and sharing his thinking openly. On being asked he thought of how Alexandria Ocasio Cortez is using Twitter, he replied that she is ‘mastering the medium’. Although Swisher managed to interview Dorsey over Twitter, the ‘Twitterview’ got quite confusing soon and went out of order. The conversations seemed all over the place and as Kurt Wagner, tech journalist from Recode puts it, “in order to find a permanent thread of the chat, you had to visit one of either Kara or Jack’s pages and continually refresh”. This made for a difficult experience overall and points towards the current flaws within the conversation system on Twitter. Many users tweeted out their opinion regarding the same: https://twitter.com/RTKumaraSwamy/status/1095542363890446336 https://twitter.com/waltmossberg/status/1095454665305739264 https://twitter.com/kayvz/status/1095472789870436352 https://twitter.com/sukienniko/status/1095520835861864448 https://twitter.com/LauraGaviriaH/status/1095641232058011648 Recode Decode #GoogleWalkout interview shows why data and evidence don’t always lead to right decisions in even the world’s most data-driven company Twitter CEO, Jack Dorsey slammed by users after a photo of him holding ‘smash Brahminical patriarchy’ poster went viral Jack Dorsey discusses the rumored ‘edit tweet’ button and tells users to stop caring about followers