6
$\begingroup$

I am asking for a book (or any other online resource) where we can solve exercises related to neural networks, similar to the books or online resources dedicated to mathematics where we can solve mathematical exercises.

$\endgroup$
  • $\begingroup$ Your question starts to look more specific! Are you looking for exercises on how to implement neural networks, or maybe are you looking for exercises that will help you understand e.g. back-propagation, etc.? What kind of exercises are you looking for? $\endgroup$ – nbro May 4 at 17:56
  • $\begingroup$ @nbro Exercises similar to the ones we have in Mathematics text book at the end of every chapter. For instance at the end of the chapter related to Calculus, there will be exercises/problems related to Calculus that we can solve. I'm looking for similar exercises related to Deep Learning, Neural Networks. $\endgroup$ – DSarkar May 4 at 18:02
  • 1
    $\begingroup$ For starters, you can go on with Udacity's deep learning with pytorch course (It's free and has exercises related to basic logistic regression using numpy to other DL applications like conv-nets, LSTM's etc in pytorch. Other than that, there is Kaggle learn(also free) and you can use kaggle competitions, even older ones for practices. Coursera has deeplerning.AI courses with coding assignments (both deep learning and tensorflow courses are great) (might be paid,not sure if auditing is available) $\endgroup$ – SajanGohil May 4 at 18:17
  • $\begingroup$ @SajanGohil Thank You. Just to clarify - are there AI courses with coding assignments in Kaggle Learn? $\endgroup$ – DSarkar May 4 at 18:23
  • 1
    $\begingroup$ I am pretty sure they make yoou do some playground stuff, but that is most likely "not from scratch", like you don't need to code backprop(other 2 do), but you can if you want $\endgroup$ – SajanGohil May 4 at 18:33
4
$\begingroup$

The book Grokking Deep Learning, by Andrew Trask (a PhD student at Oxford University and a research scientist at DeepMind), a wonderful, clean, and plain-English discussion of the basic mechanics that go on under the hood of neural networks - from data flow to updating of weights. It is written without a slant on normally-wonky math, the concepts are presented and then advanced at a digestible pace for anyone.

Here are a few more possibly useful resources.

  1. Neural Networks: Playground Exercises

  2. Getting Started With Deep Learning: Convolutional Neural Networks

  3. Deep learning focuses on practical aspects of deep learning

  4. First lab assignment in Deep learning

  5. Deeplearning stanford

| improve this answer | |
$\endgroup$
4
$\begingroup$

There are actually quite a few. Personally I would say these courses have high quality and strong focus on practice:

  • Standford computer vision cs231. Check the assignments materials on this page. This course has good explanation/exercises of how generally neural nets and backprop works.
  • Fastai course notebooks. You can listen to the lectures as well, but notebooks are quite self-containing
  • Practical reinforcement learning course, if you interested in NN application for RL
| improve this answer | |
$\endgroup$
  • $\begingroup$ The question is about neural networks, but you're listing resources that aren't primarily related to neural networks!! $\endgroup$ – nbro May 4 at 22:54
  • 1
    $\begingroup$ I would say they are primarily related. I.e. in cs231 80% of assignments are directly about neural nets and other 20% helps to understand them better (like understanding how optimization works on svm in fact help to understand how optimizers work for NN as wel). $\endgroup$ – Kirill Fedyanin May 4 at 23:06
  • $\begingroup$ I don't really think so. Those resources are not primarily dedicated to neural networks. Neural networks aren't just convolutional or feed-forward neural networks. There are many other neural networks. $\endgroup$ – nbro May 5 at 11:59
  • 1
    $\begingroup$ @nbro In my opinion, the CS231n course is by far the most useful resource to learn about neural networks in general - it's not only about convnets. The version taught by Andrej Karpathy is truly extraordinary. $\endgroup$ – Mathias Müller May 6 at 7:32
3
$\begingroup$

One of the most famous books dedicated to neural networks is Neural Networks - A Systematic Introduction (1996) by Raul Rojas. Most chapters end with a series of exercises that test your understanding of the material. For example, in chapter 14 Stochastic Networks, one of the exercises is

Solve the eight queens problem using a Boltzmann machine. Define the network's weights by hand.

This should give you a sense of the type of exercise that you will find in this book.

| improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.