54
3

Efficient and Optimal Fixed-Time Regret with Two Experts

Abstract

Prediction with expert advice is a foundational problem in online learning. In instances with TT rounds and nn experts, the classical Multiplicative Weights Update method suffers at most (T/2)lnn\sqrt{(T/2)\ln n} regret when TT is known beforehand. Moreover, this is asymptotically optimal when both TT and nn grow to infinity. However, when the number of experts nn is small/fixed, algorithms with better regret guarantees exist. Cover showed in 1967 a dynamic programming algorithm for the two-experts problem restricted to {0,1}\{0,1\} costs that suffers at most T/2π+O(1)\sqrt{T/2\pi} + O(1) regret with O(T2)O(T^2) pre-processing time. In this work, we propose an optimal algorithm for prediction with two experts' advice that works even for costs in [0,1][0,1] and with O(1)O(1) processing time per turn. Our algorithm builds up on recent work on the experts problem based on techniques and tools from stochastic calculus.

View on arXiv
Comments on this paper