Artificial Intelligence Stack Exchange is a question and answer site for people interested in conceptual questions about life and challenges in a world where "cognitive" functions can be mimicked in purely digital environment. It only takes a minute to sign up.
Sign up to join this community
I'm studying a master's degree and my final work is going to be about the convolutional neural network.
I read a lot of books and I did Convolutional Network Standford's course, but I need more.
Are there books or papers on the details of convolutional neural networks (in particular, convolutional layer)?
Chapter 9 of the book Deep Learning (2016), by Goodfellow et al., describes the convolutional (neural) network (CNN), its main operations (namely, convolution and pooling) and properties (such as parameter sharing).
There's also the article From Convolution to Neural Network, which first introduces the mathematical operation convolution and then describes its connection with signal processing (where images can be viewed as 2D signals) and, finally, describes the CNN.
Chris Olah's work is always inspired, and not too technical as one would expect. He has several papers on CNNs on his website. In particular, check the series titled "Convolutional Neural Networks" with four papers on the topic.
You can look at the paper Gradient-Based Learning Applied to Document Recognition (1998) by Yann LeCun et al., which reviews and compares various methods applied to handwritten character recognition and shows that CNNs outperform all other methods.
Also, I suggest Andrew Ng's CNN videos.
I'm not sure if this is what you are looking for but I find Goodfellow's book a pretty good resource:
Goodfellow, specifically Section 2, Chapter 9 deals with convolutional neural networks: https://www.deeplearningbook.org/
'Pattern Recognition and Machine Learning' by Bishop Might contains a section (5.5.5, pg 267 onwards) as well as an exercise, and a general discussion about neural networks in image recognition.
If you edit your question to post a bit more detail, we can offer better answers, for example, what is about the convolutional layer? How it's implemented?
If you are looking for a more basic introduction to convolutional layers I would also suggest:
A Comprehensive Guide to Convolutional Neural Networks — the ELI5 way gives a pretty general overview, starting at the difference between CNNs and ANNs and explains why CNNs are superior to ANNs (for certain problems). It also gives some details about how the convolution actually works.
Demystifying the transpose convolution explains the transpose convolution operation in the context of how a traditional convolution; this may not be relevant if you are strictly using CNNs and not transpose-CNNs.
Understanding of Convolutional Neural Network (CNN) — Deep Learning is quite similar to "A Comprehensive..." link above, but it also includes information about filtering and shows the effect that different filters have on an image, which is certainly very import to an understanding of why we use CNNs.
Building a Convolutional Neural Network (CNN) in Keras (or one of the other thousand similar pages) are pretty good for just starting out and building your own CNN classifier. You can also check out examples from Keras, e.g. CIFAR10 CNN, but these tend to give you a very little information about why they designed the network the way that they did.
If, on the other hand, you are looking for some more advanced resources, here are is one that springs to mind:
Deep Residual Learning for Image Recognition by He et al., deals with a major advance in image recognition, using Residual Networks (ResNet). This type of network has become pretty popular, so I highly recommend giving it a read.
