Programme‎ > ‎


Introduction to Independent Component Analysis

Arthur Gretton

Independent component analysis (ICA) is a technique for extracting underlying sources of information from linear mixtures of these sources, based only on the assumption that the sources are independent of each other. To illustrate the idea, we might be in a room containing several people (the sources) talking simultaneously, with microphones picking up multiple conversations at once (the mixtures), and we might wish to automatically recover the original separate conversations from these mixtures. More broadly, ICA is used in a very wide variety of applications, including signal extraction from EEG, image processing, bioinformatics, and economics. I will present an introduction to ICA, which includes a description of principal component analysis (PCA), and how ICA differs from PCA, the maximum likelihood approach, the case where fixed nonlinearities are used as heuristics for source extraction, some more modern information theoretic approaches, and a kernel-based method. I will also cover two optimization strategies, and provide a comparison of the various approaches on benchmark data, to reveal the strengths and failure modes of different ICA algorithms (with a focus on modern, recently published methods).

Learning and Inference in Vision: from Features to Scene Understanding

Jonathan Huang, Tomasz Malisiewicz

Recent advances in Computer Vision and Machine Learning, coupled with the availability of large labeled image datasets as well as new and cheap sensor technologies, have spurred interest in object recognition and scene understanding.  In this tutorial we focus on several recent research trends in this area.  We plan to cover recent work on a.) learning useful low-level feature representations for vision tasks, b.) learning to combine such features for part-based object detection, and finally c.) learning and reasoning with context for producing semantically coherent understanding of scenes.
Our emphasis will be on illuminating the important learning and inference problems that have arisen in computer vision.
Hai-Son Le,
Nov 29, 2009, 3:08 PM