Contextual Multi-armed Bandit Algorithm for Semiparametric Reward Model

TL;DR

This paper introduces a new contextual multi-armed bandit algorithm for semiparametric reward models that handles nonstationarity, offering improved simplicity and speed while maintaining optimal regret bounds.

Contribution

It presents a less restrictive, faster algorithm for semiparametric reward models with nonstationarity, achieving regret bounds comparable to linear models.

Findings

Algorithm outperforms existing methods in simulations.

Effective in real-world news recommendation data.

Achieves tight regret upper bounds similar to Thompson sampling.

Abstract

Contextual multi-armed bandit (MAB) algorithms have been shown promising for maximizing cumulative rewards in sequential decision tasks such as news article recommendation systems, web page ad placement algorithms, and mobile health. However, most of the proposed contextual MAB algorithms assume linear relationships between the reward and the context of the action. This paper proposes a new contextual MAB algorithm for a relaxed, semiparametric reward model that supports nonstationarity. The proposed method is less restrictive, easier to implement and faster than two alternative algorithms that consider the same model, while achieving a tight regret upper bound. We prove that the high-probability upper bound of the regret incurred by the proposed algorithm has the same order as the Thompson sampling algorithm for linear reward models. The proposed and existing algorithms are evaluated via simulation and also applied to Yahoo! news article recommendation log data.