Oral
Semiparametric Contextual Bandits
Akshay Krishnamurthy · Steven Wu · Vasilis Syrgkanis

Fri Jul 13th 11:20 -- 11:40 AM @ A5

This paper studies semiparametric contextual bandits, a generalization of the linear stochastic bandit problem where the reward for a chosen action is modeled as a linear function of known action features confounded by a non-linear action-independent term. We design new algorithms that achieve $\tilde{O}(d\sqrt{T})$ regret over $T$ rounds, when the linear function is $d$-dimensional, which matches the best known bounds for the simpler unconfounded case and improves on a recent result of Greenwald et al. (2017). Via an empirical evaluation, we show that our algorithms outperform prior approaches when there are non-linear confounding effects on the rewards. Technically, our algorithms use a new reward estimator inspired by doubly-robust approaches and our proofs require new concentration inequalities for self-normalized martingales.

Author Information

Akshay Krishnamurthy (Microsoft Research)
Steven Wu (Microsoft Research)
Vasilis Syrgkanis (Microsoft Research)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors