Chapter 10: Comet for Deep Learning
Deep learning is a subfield of artificial intelligence that aims to extract knowledge from data through complex neural networks. You can imagine a neural network as a collection of nodes (or neurons), which are organized in different layers of processing. Similar to the human brain, which learns in incremental steps, in a neural network, the level of knowledge increases as you move from one layer to the next.
Compared to traditional machine learning algorithms, deep learning algorithms can automatically extract features from incoming data. Therefore, they are able to process even thousands of input features, which is unthinkable for a machine learning algorithm in which you should select the input features by hand.
Running deep learning algorithms is both computationally expensive and resource-consuming. However, over the last few years, their uptake has increased considerably thanks to the development of even more powerful machines as well...
