2020/02/17 by Thibaut Cuvelier, Cuvelier, Thibaut, Richard Combes +3 · 1 citation
Decision Sciences · Computer Science · #Advanced Bandit Algorithms Research #Auction Theory and Applications #Machine Learning and Algorithms
paper · doi:10.48550/arxiv.2002.07258
We consider combinatorial semi-bandits over a set of arms \cal X ⊂ \0,1\d where rewards are uncorrelated across items. For this problem, the algorithm ESCB yields the smallest known regret bound R(T) = \cal O( d (ln m)2 (ln T) \over Δmin ), but it has computational complexity \cal O(|\cal X|) which is typically exponential in d, and cannot be used in large dimensions. We propose the first algorithm which is both computationally and statistically efficient for this problem with regret R(T) = \cal O (d (ln m)2 (ln T)\over Δmin ) and computational complexity \cal O(T \bf poly(d)). Our approach involves carefully designing an approximate version of ESCB with the same regret guarantees, showing that this approximate algorithm can be implemented in time \cal O(T \bf poly(d)) by repeatedly maximizing a linear function over \cal X subject to a linear budget constraint, and showing how to solve this maximization problems efficiently.