We consider a multi-armed bandit problem where payoffs are a linear function of an observed stochastic contextual variable. In the scenario where there exists a gap between optimal and suboptimal rewards, several algorithms have been proposed that achieve O(logT) regret after T time steps. However, proposed methods either have a computation complexity per iteration that scales linearly with T or achieve regrets that grow linearly with the number of contexts |χ|. We propose an ε-greedy type of algorithm that solves both limitations. In particular, when contexts are variables in ℝd, we prove that our algorithm has a constant computation complexity per iteration of O(poly(d)) and can achieve a regret of O(poly(d) log T) even when |χ| = Ω(2d). In addition, unlike previous algorithms, its space complexity scales like O(Kd2) and does not grow with T. © 2013 Springer-Verlag.
CITATION STYLE
Bento, J., Ioannidis, S., Muthukrishnan, S., & Yan, J. (2013). A time and space efficient algorithm for contextual linear bandits. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8188 LNAI, pp. 257–272). https://doi.org/10.1007/978-3-642-40988-2_17
Mendeley helps you to discover research relevant for your work.