reading lists for new lisa students 1 from bengio

info coming from bengio

Research in General

[1] How to write a great research paper

Basics of deep learning

[1] Learning deep architectures for AI

[2] Practical recommendations for gradient-based training of deep architectures

[3] Quick’n’dirty introduction to deep learning: Advances in Deep Learning

[4] A fast learning algorithm for deep belief nets

[5] Greedy Layer-Wise Training of Deep Networks

[6] Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion

[7] Contractive auto-encoders: Explicit invariance during feature extraction

[8] Why does unsupervised pre-training help deep learning?

[9] An Analysis of Single Layer Networks in Unsupervised Feature Learning

[10] The importance of Encoding Versus Training With Sparse Coding and Vector Quantization

[11] Representation Learning: A Review and New Perspectives

[12] Deep Learning of Representations: Looking Forward

[13] Measuring Invariances in Deep Networks

[14] Neural networks course at USherbrooke [youtube]

Feedforward nets

[1] “Improving Neural Nets with Dropout” by Nitish Srivastava

[2] “Deep Sparse Rectifier Neural Networks”

[3] “What is the best multi-stage architecture for object recognition?”

[4] “Maxout Networks”

MCMC

[1] Iain Murray’s MLSS slides

[2] Radford Neal’s Review Paper (old but still very comprehensive)

[3] Better Mixing via Deep Representations