christoschristofidis/awesome-deep-learning

Awesome Deep Learning is a curated list of resources for people who want to study deep learning, the branch of machine learning that uses neural networks. The list is maintained on GitHub by Christos Christofidis, follows the well-known Awesome-list convention, and points readers to outside material rather than hosting any teaching content itself. The contents are organized into a long table of links, grouped into sections: books, courses, video lectures, papers, tutorials, researchers, websites, datasets, conferences, frameworks, tools, and a miscellaneous bucket. The books section lists titles such as Goodfellow, Bengio and Courville's Deep Learning, Michael Nielsen's Neural Networks and Deep Learning, the interactive d2l.ai textbook, and a long run of Manning press books on more specific topics like JAX, evolutionary methods, regularization, TensorFlow, and deep learning for natural language processing. The courses section is the longest portion shown. It links to university lectures from Stanford, including Andrew Ng's original Coursera machine learning class, the convolutional networks class taught by Fei-Fei Li and Andrej Karpathy, and the natural language class from CS224d. It also points to courses at Caltech, Carnegie Mellon, MIT, Berkeley, Oxford, Waterloo, and the University of Amsterdam, plus free industry offerings like Udacity's Deep Learning with Google, Fast.ai's Practical Deep Learning for Coders, and OpenAI's Spinning Up in Deep Reinforcement Learning. A few entries are short bootcamps, others are full graduate-level sequences. The purpose of the repo is bookmarking. Someone new to the field can open the table of contents, pick a format that suits them (book, course, paper, framework documentation), and follow the link out to the source. The README does not rank entries, does not recommend a learning path, and does not provide any code of its own. Readers choose based on their own background. Because it is a list, the repo has no install steps and nothing to run, it is a static document of links, kept current through pull requests from contributors.