Introduction

At the Colorado State Fair in 2022, the winning entry was a fantastical sci-fi landscape created by video game designer Jason Allen titled Théâtre D’opéra Spatial (Figure 1.1). The first prize art was remarkable both from the dramatic subject matter, but also due to the unusual origin of this image. Unlike the majority of other artworks entered into the competition, the Théâtre D’opéra Spatial was not painted using oil or watercolors, nor was its “creator” even human; rather, it is an entirely digital image produced by a sophisticated machine learning algorithm called Midjourney. Jason used Midjourney, which has been trained on diverse images, along with natural language instructions to create the image, rather than a brush and canvas.

Visual art is far from the only area in which machine learning has demonstrated astonishing results. Indeed, if you...

The rules of probability

At the simplest level, a model, be it machine learning or a more classical method such as linear regression, is a mathematical description of how a target variable changes in response to variation in a predictive variable; that relationship could be a linear slope or any of a number of more complex mathematical transformations. In the task of modeling, we usually think of separating the variables in our dataset into two broad classes:

• Independent data, by which we primarily mean inputs to a model, is often denoted by X. For example, if we are trying to predict the grades of school students on an end-of-year exam based on their characteristics, we could think of several kinds of features:
  • Categorical: If there are six schools in a district, the school that a student attends could be represented by a six-element vector for each student. The elements are all zero, except for one which is “1,” indicating which of the six schools they are enrolled in...

Discriminative and generative modeling, and Bayes’ theorem

Now, let us consider how these rules of conditional and joint probability relate to the kinds of predictive models that we build for various machine learning applications. In most cases—such as predicting whether an email is fraudulent or the dollar amount of the future lifetime value of a customer—we are interested in the conditional probability, P(Y|X=x), where Y is the set of outcomes we are trying to model and X is the input “features,” and x is a particular value of the input features. For example, we are trying to calculate the probability that an email is fraudulent based on the knowledge of the set of words (the x) in the message. This approach is known as discriminative modeling (Ref 15-17). Discriminative modeling attempts to learn a direct mapping between the data, X, and the outcomes, Y.

Another way to understand discriminative modeling is in the context of Bayes’ theorem ...

Why generative models?

Now that we have reviewed what generative models are and defined them more formally in the language of probability, why would we have a need for such models in the first place? What value do they provide in practical applications? To answer this question, let us take a brief tour of the topics that we will cover in more detail in the rest of this book.

Unique challenges of generative models

Given the powerful applications that generative models are applied to, what are the major challenges in implementing them? As described, most of these models utilize complex data, requiring us to fit large models with sufficiently diverse inputs to capture all the nuances of their features and distribution. That complexity arises from sources including:

• Range of variation: The number of potential images generated from a set of three color channel pixels is immense, as is the vocabulary of many languages
• Heterogeneity of sources: Language models, in particular, are often developed using a mixture of data from several websites
• Size: Once data becomes large, it becomes more difficult to catch duplications, factual errors (such as mistranslations), noise (such as scrambled images), and systematic biases
• Rate of change: Many developers of LLMs struggle to keep model information current with the state of the world and thus provide relevant answers to...

Summary

In this chapter, we discussed what generative modeling is, and how it fits into the landscape of more familiar machine learning methods, using probability theory and Bayes’ theorem to describe how these models approach prediction in an opposite manner to discriminative learning. We reviewed use cases for generative learning, both for specific kinds of data and general prediction tasks. As we saw, text and images are the two major forms of data that these models are applied to. For images, the major models we will discuss are VAE, GAN, and similar algorithms. For text, the dominant models are transformer architectures such as Llama, GPT, and BERT. Finally, we examined some of the specialized challenges that arise from building these models.

References

1. https://www.smithsonianmag.com/smart-news/artificial-intelligence-art-wins-colorado-state-fair-180980703/
2. ChatGPT technical report: https://arxiv.org/abs/2303.08774
3. https://arxiv.org/abs/2107.03374
4. https://www.nature.com/articles/s41598-019-42294-8
5. https://deepmind.com/research/case-studies/alphago-the-story-so-far
6. https://deepmind.com/blog/article/AlphaStar-Grandmaster-level-in-StarCraft-II-using-multi-agent-reinforcement-learning
7. https://arxiv.org/abs/1810.04805
8. Brandon, John (16 February 2018). Terrifying high-tech porn: Creepy ‘deepfake’ videos are on the rise, Fox News. Retrieved 20 February 2018.
9. https://upload.wikimedia.org/wikipedia/en/thumb/7/71/Deepfake_example.gif/280px-Deepfake_example.gif
10. https://devopstar.com/static/2293f764e1538f357dd1c63035ab25b0/d024a/fake-facebook-conversation-example-1.png
11. https://openai.com/blog/better-language-models/
12. https://ai.googleblog.com/2018/05/duplex-ai-system-for-natural-conversation.html
13. https://salu133445.github...