These are some of the very recent and fairly advanced works in some of the leading research areas in theoretical deep learning. For some of the more introductory and application oriented resources check out our Computer Vision summer school
Resources on Theoretical Deep Learning
Deep Learning Theory, Talks, Tutorials, Blogs and Courses
Deep Learning Theoretical Papers
Theory of neural networks
- Numerics of GANs, NIPS 2017
- GANs are Broken in More than One Way, explained by Ferenc Huszar
- Generalisation and Equilibrium in GANs, by Sanjeev Arora
- Improved Variational Inference with Inverse Autoregressive Flow Diederik P. Kingma, Tim SalimansTim Salimans, Rafal Jozefowicz, Xi Chen, Max Welling, NIPS 2016.
- Improved Techniques for Training GANs Tim Salimans, Ian Goodfellow, Wojciech Zaremba, Vicki Cheung, Alec Radford, Xi Chen, arxiv 2016.
Bayesian deep learning
- Variational Information MaximizingExploration Rein Houthooft, Xi Chen, Yan Duan John Schulman , Filip De Turck Pieter Abbeel, arXiv 2017.
- Reinforcement learning of motor skills with policy gradientsJan Peters, Stefan Schaal, Neural Networks 2008.
- Continuous Control With Deep Reinforcement Learning Timothy P. Lillicrap, Jonathan J. Hunt, Alexander Pritzel, Nicolas Heess, Tom Erez, Yuval Tassa, David Silver, ICLR 2016.