ICCV 2003 Short Course

Learning and Vision: Discriminative Methods

Instructors: Christopher Bishop (Microsoft Research Limited) and Paul Viola (Microsoft Research).

Duration: 3.5 hours

Course Content

1. Foundations:
   • Inference and decision
   • Discriminative versus generative approaches to classification and regression
   • Bayesian methods
   • Feature selection
2. Algorithms, architectures, and techniques:
   • Multilayer neural networks
   • Gaussian processes
   • Support vector machines
   • Decision trees
   • Adaboost
   • Memory based classification/estimation and lcoally-weighted regression
   • Estimating generalization: Cross-validation and ROC curves
   • Building fast classifiers
3. Applications:
   • Face and object detection
   • Recognition: Characters/faces/objects
   • Body pose estimation

Biography

Chris Bishop graduated from Oxford with B.A. in Physics, and obtained a PhD in Theoretical Physics from the University of Edinburgh a thesis on quantum field theory. After several years working on the theory of magnetically confined plasmas for the fusion programme, he developed an interest in pattern recognition, and became Head of the Applied Neurocomputing Centre at AEA Technology. In 1993 he was elected to a Chair in the Department of Computer Science and Applied Mathematics at Aston University, where he was a member of the Neural Computing Research Group. He then took a sabbatical to be principal organiser of the six month international research programme on Neural Networks and Machine Learning at the Isaac Newton Institute for Mathematical Sciences in Cambridge, which ran from July to December 1997.

After completion of the Newton Institute programme he joined the Microsoft Research Laboratory in Cambridge, where he is Assistant Director. At the same time he was elected to a Chair of Computer Science at the University of Edinburgh where he is a member of the Institute for Adaptive and Neural Computation in the Division of Informatics. He is also a Fellow of Darwin College, Cambridge.

His research is concerned with the development and application of probabilistic methods for inference and learning.

Before moving to Microsoft, Paul Viola was a researcher at MERL and an Associate Professor of Computer Science and Engineering at the Massachusetts Institute of Technology. He also spent two years as a visiting scientist in the Computational Neurobiology of the Salk Institute in San Diego. Paul has a broad background in advanced computational techniques, publishing in the fields of computer vision, machine learning, medical imaging, neurobiological vision, mobile robotics, and automated drug design. Paul was a recipient of a National Science Foundation Career award in 1998.