Useful references
Books
- Bayesian Reasoning and Machine Learning by David Barber [pdf]
- The Elements of Statistical Learning by Trevor Hastie, Robert Tibshirani and Jerome Friedman [html] [pdf]
- Information Theory, Inference, and Learning Algorithms by David J.C. MacKay [html] [pdf]
- Convex Optimization
by Stephen Boyd and Lieven Vandenberghe [pdf] [html]
- Natural Image Statistics by Aapo Hyvärinen, Jarmo Hurri and Patrik O. Hoyer [html] [pdf]
- The Quest for Artificial Intelligence - A History of Ideas and Achievements by Nils J. Nilsson [html] [pdf]
- Gaussian Processes for Machine Learning by Carl Edward Rasmussen and Christopher K. I. Williams [html] [pdf]
- Introduction to Information Retrieval by Christopher D. Manning, Prabhakar Raghavan and Hinrich Schütze [html] [pdf]
Journals
- Journal of Machine Learning Research [html] [papers]
- Neural Computation [html]
Conferences
- Neural Information Processing Systems (NIPS) [html] [papers]
- International Conference on Machine Learning (ICML) [html]
- International Conference on Artificial Intelligence and Statistics (AISTATS) [html]
- Uncertainty in Artificial Intelligence (UAI) [html] [papers]
Others
- Learning deep architectures for AI (literature review on deep learning) by Yoshua Bengio [pdf]
- The Matrix Cookbook by Kaare Brandt Petersen and Michael Syskind Pedersen [pdf]
- Structured Learning and Prediction in Computer Vision (tutorial on structured output prediction applied to computer vision) by Sebastian Nowozin and Christoph Lampert [pdf]
Code and datasets
- MLPython: my group's machine learing research library (see the documentation here);
- Theano:
a Python library for easily defining, computing,
optimizing and symbolicallly manipulating mathematical
expressions, on the CPU or GPU;
- CUDAMat:
a Python library supporting the computation of common
matrix operations on the GPU;
- GNumPy:
a NumPy-like library for easily manipulating
matrices on the GPU;
- deeplearning.net:
a website dedicated to deep learning, that references many datasets and libraries
useful in deep learning research;
- LIBSVM datasets: a list of datasets for machine learning research, all in the LIBSVM format.