Improved second-order bounds for prediction with expert advice
- 655 Downloads
This work studies external regret in sequential prediction games with both positive and negative payoffs. External regret measures the difference between the payoff obtained by the forecasting strategy and the payoff of the best action. In this setting, we derive new and sharper regret bounds for the well-known exponentially weighted average forecaster and for a second forecaster with a different multiplicative update rule. Our analysis has two main advantages: first, no preliminary knowledge about the payoff sequence is needed, not even its range; second, our bounds are expressed in terms of sums of squared payoffs, replacing larger first-order quantities appearing in previous bounds. In addition, our most refined bounds have the natural and desirable property of being stable under rescalings and general translations of the payoff sequence.
KeywordsIndividual sequences Prediction with expert advice Exponentially weighted averages
- Allenberg-Neeman, C., & Neeman B. (2004). Full information game with gains and losses. Algorithmic Learning Theory, 15th International Conference, ALT 2004, Padova, Italy, October 2004, In Proceedings, volume 3244 of Lecture Notes in Artificial Intelligence, pp. 264-278, Springer.Google Scholar
- Cesa-Bianchi, N., & Lugosi, G. (2006). Prediction, Learning, and Games. Cambridge University Press.Google Scholar
- Helmbold, D.P., Schapire, R.E., Singer, Y., & Warmuth, M. K. (1998). On-line portfolio selection using multiplicative updates. Mathematical Finance, 8,325–344, 1998.Google Scholar
- Piccolboni, A., & Schindelhauer, C. (2001). Discrete prediction games with arbitrary feedback and loss. In Proceedings of the 14th Annual Conference on Computational Learning Theory (pp. 208–223).Google Scholar