DeepRead
Cover of Generative Deep Learning

Generative Deep Learning

David Foster · 2022

Machine learning, statistics and probability › Deep learning and AI

#7,770 most recommended

About the book

A practical introduction to generative deep learning: models that learn the probability distribution of a training dataset and then sample new data from it, such as images, text, music and games. It covers the main model families, including variational autoencoders, generative adversarial networks, autoregressive models and Transformers, and it builds each one with code in TensorFlow and Keras. The author's central position is that a range of generative methods share one underlying idea, learning a distribution over data, so they can be understood through a common framework rather than as unrelated techniques. The book claims that the rapid progress in generative models comes from a small set of architectural and training ideas that a practitioner can implement and experiment with directly. It also presents generative modelling as a way to study how a model represents the structure of its data, not only as a way to produce output.

Written by Claude from what it knows of this book.

What Hacker News says 25% recommend · 0% critical

024202420252025 peak · 3
This bookAverage book in Deep learning and AI

In the commenters’ own words

What Open Library readers say

No Open Library reader has rated it yet.

  • 18 want to read
  • 2 reading
  • 0 have read

Open Library’s page for the book

What critics say, via Wikipedia

No Wikipedia article about this book yet, so no critics to quote.

Reddit communities — coming later

Details

First published
2022
ISBN
9781098134181
Editions known
6 at Open Library
Open Library
OL28799243W

Explore in the mosaic

Read next: its neighbours on the mosaic