Multi-Task Off-Policy Learning from Bandit Feedback

TL;DR

This paper introduces HierOPO, a hierarchical off-policy optimization method for multi-task bandit problems, demonstrating improved policy learning by leveraging shared structure among tasks through theoretical bounds and empirical evaluation.

Contribution

The paper presents HierOPO, a novel hierarchical off-policy optimization algorithm for multi-task bandit feedback, with theoretical guarantees and efficient implementation for linear Gaussian models.

Findings

HierOPO outperforms non-hierarchical methods in policy quality.

Theoretical bounds show improved suboptimality with hierarchy.

Empirical results confirm the advantage of shared structure in tasks.

Abstract

Many practical applications, such as recommender systems and learning to rank, involve solving multiple similar tasks. One example is learning of recommendation policies for users with similar movie preferences, where the users may still rank the individual movies slightly differently. Such tasks can be organized in a hierarchy, where similar tasks are related through a shared structure. In this work, we formulate this problem as a contextual off-policy optimization in a hierarchical graphical model from logged bandit feedback. To solve the problem, we propose a hierarchical off-policy optimization algorithm (HierOPO), which estimates the parameters of the hierarchical model and then acts pessimistically with respect to them. We instantiate HierOPO in linear Gaussian models, for which we also provide an efficient implementation and analysis. We prove per-task bounds on the suboptimality of the learned policies, which show a clear improvement over not using the hierarchical model. We also evaluate the policies empirically. Our theoretical and empirical results show a clear advantage of using the hierarchy over solving each task independently.