By now you’ve probably familiarized yourself with the convolutional neural network that won the ImageNet challenge in 2012. To see how it has evolved, check out the first paper that I’ve linked below, where researchers from google expanded the typical CNN architecture and created what they called “Inception”.
Recurrent neural networks are the next topic. Check out the blog first, and then dive into the second paper that I’ve linked.
“Understanding LSTM networks”
“Going deeper with convolutions”
“Sequence to Sequence Learning with Neural Networks”