Minimax Square Loss Prediction
The squared Euclidean distance is one of the widely used loss functions in machine learning. We study prediction games with square loss using a pure minimax approach. The goal is to obtain efficient algorithms for natural tasks.
-
Minimax time series prediction
Wouter M. Koolen, Alan Malek, Peter L. Bartlett, Yasin Abbasi-Yadkori
In Advances in Neural Information Processing Systems (NeurIPS) 28, pages 2548–2556, December 2015.
-
Minimax fixed-design linear regression
Peter L. Bartlett, Wouter M. Koolen, Alan Malek, Manfred K. Warmuth, Eiji Takimoto
In Proceedings of the 28th Annual Conference on Learning Theory (COLT), pages 226–239, June 2015.
-
Efficient minimax strategies for square loss games
Wouter M. Koolen, Alan Malek, Peter L. Bartlett
In Advances in Neural Information Processing Systems (NeurIPS) 27, pages 3230–3238, December 2014.