In addition to basic forward networks, many books cover other basic network designs like CNN's and RNN's. However, they don't really go any further than that, explaining things like common approaches to solve complex tasks with more complex systems, e.g. dividing the work/data between networks, chaining different types of networks for different tasks and combining different partial solutions to a final solution.

For example, I read that it's common to use multiple RNN's in serial and/or parallel for language processing like translating text, dividing the work to something like understanding the intent first, then translating it, avoiding a (probably wrong) literal translation of the plain words without any meaning. Another example I remember reading about is how game AI's, e.g. a chess AI, use different types of networks to analyze moves, weight its options and decide on strategies. Or a robot using a CNN to analyse camera footage and then using other networks to process the input and decide on what action to take.

I found this post, which mentions to books. I know the first one (NNDesign), though I haven't read it yet, but it doesn't seem to cover what I'm looking for. The second one seems to have a chapter on modular networks, which seems to go into the right direction. It's now much, though.

Does anyone have some good books/resources on this kind of topics?


1 Answer 1


"Deep Learning" by Ian Goodfellow, Yoshua Bengio, and Aaron Courville

"Neural Network Methods for Natural Language Processing" by Yoav Goldberg

"Deep Reinforcement Learning" by Pieter Abbeel and John Schulman

  • $\begingroup$ Thank you for your reply. I will take a look at these books. $\endgroup$ Commented Dec 15, 2023 at 13:28
  • $\begingroup$ @ThorstenSchmitz This user has used ChatGPT or a similar model to generate this and other answers. I'd not trust him or his answers. In fact, the first book mentioned does not really cover in depth complex neural networks. The others I don't know. $\endgroup$
    – nbro
    Commented Dec 28, 2023 at 0:44

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .