Getting Started with Recurrent Neural Networks (RNNs) and Convolutional Neural Networks
This page is intended for people who have some background in Machine Learning and want to learn more about Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs) and apply them to their datasets.
LKS-CHART projects that apply neural nets include An Early Warning System for General Internal Medicine and Automated Analysis of Cervical Spine CT Scans in Trauma.
Learning about RNNs:
Chapter 9 of Goodfellow et al.'s definitive text book Deep Learning provides a good overview of RNNs.
People new to the field of RNNs might benefit more from the following:
Andrej Karpathy's lecture CS231n Winter 2016: Lecture 10: Recurrent Neural Networks, Image Captioning, LSTM from his CS231n at Stanford provides a quick overview of RNNs.
Richard Socher's CS224d: Deep Learning for Natural Language Processing, also at Stanford, provides a comprehensive overview of RNNs, with a focus of applying RNNs to Natural Language Processing (NLP).
Getting Started with RNNs: Warm-up
We recommend the introductory assignment Understanding How Recurrent Neural Networks Model Text developed at the University of Toronto, as well as the relevant assignments from CS231n and CS224d.
Getting Started with RNNs: Applying RNNs to your own dataset
You probably do not want to code your LSTM RNN from scratch! A lot of ready-made code is available here https://github.com/ilkarman/DeepLearningFrameworks
Convolutional Neural Networks
Stanford's CS231n is an up-to-date introduction to Convolutional Neural Networks, and is updated every year.
Goodfellow et al.'s Deep Learning provides a definitive introduction to deep learning, including Convolutional Neural Networks.
Michael Nielsen's Neural Networks and Deep Learning provides a gentler introduction to neural networks.
Programming Deep Neural Network Systems
Deep Learning with R by Francois Chollet with J.J Allaire is a great overview of deep learning with applications in R. Examples use the Keras library for R.
A curated list of awesome machine learning by Joseph Misiti
Machine Learning Yearning by Andrew Ng is a free book that teaches you how to structure Machine Learning projects.
Continue to explore