Online Learning, Sequential Prediction, Regret Minimization
Investigators: Alexander (Sasha) Rakhlin
In this project we study sequential prediction methods. Particular areas of research are
- Relation to statistical (batch) learning
- Notions of complexity for online prediction
- Minimax analysis
- Application of sequential prediction methods to aggregation methods and martingale arguments
Recent papers on the subject:
J. Abernethy, A. Agarwal, P. Bartlett, and A. Rakhlin. A Stochastic View of Optimal Regret through Minimax Duality, COLT 2009.
J. Abernethy, P. Bartlett, A. Rakhlin, and A. Tewari. Optimal Strategies and Minimax Lower Bounds for Online Convex Games, COLT 2008.
P. Bartlett, E. Hazan, and A. Rakhlin. Adaptive Online Gradient Descent, NIPS 2007.
J. Abernethy, P. Bartlett, and A. Rakhlin. Multitask Learning with Expert Advice, COLT 2007.
A. Rakhlin, J. Abernethy, and P. Bartlett. Online Discovery of Similarity Mappings , ICML 2007.