GCB'06 Kernel Methods Tutorial
Tutorial on "Kernel Methods for Predictive Sequence Analysis" at GCB 2006 in Tübingen by Cheng Soon Ong and Gunnar Rätsch
Cheng Soon Ong and Gunnar Rätsch will give a tutorial on Kernel Methods for Predictive Sequence Analysis at the German Conference on Bioinformatics.
The tutorial is meant for a broad audience: Students, researchers, biologists and computer scientist interested in (a) an overview of general and efficient algorithms for statistical learning used in computational biology, (b) sequence kernels for the problems such as promoter or splice site detection. No specific knowledge will be required since the tutorial is self-contained and most fundamental concepts are introduced during the course.
Overview:
- Machine learning & support vector machines
- Kernels
- Basics
- Substring kernels (Spectrum, WD, . . . )
- Efficient data structures
- Other kernels (Fisher Kernel, . . . )
- Some theoretical aspects
- Margins & Complexity Control
- Model Selection
- Loss functions & Regularization
- Regression & Multi-Class problems
- Representer Theorem
- Extensions
- Structure Learning
- Multiple Kernel Learning
- Applications
Online Material:
- Summary
- Slides
- Kernel Learning & Sequence analysis toolbox Shogun
- Simulation examples (Matlab code and data)