Hyperparameter Optimization Can Even be Harmful in Off-Policy Learning and How to Deal with It

TL;DR

This paper reveals that standard hyperparameter optimization methods can be harmful in off-policy learning due to overestimation issues, and proposes simple corrections to improve reliability.

Contribution

It identifies the pitfalls of naive HPO in off-policy learning and introduces effective, computationally efficient correction methods to address overestimation problems.

Findings

Naive HPO can lead to selecting hyperparameters with overestimated performance.

Proposed corrections improve HPO reliability in off-policy learning.

Empirical results show the effectiveness of the corrected HPO method.

Abstract

There has been a growing interest in off-policy evaluation in the literature such as recommender systems and personalized medicine. We have so far seen significant progress in developing estimators aimed at accurately estimating the effectiveness of counterfactual policies based on biased logged data. However, there are many cases where those estimators are used not only to evaluate the value of decision making policies but also to search for the best hyperparameters from a large candidate space. This work explores the latter hyperparameter optimization (HPO) task for off-policy learning. We empirically show that naively applying an unbiased estimator of the generalization performance as a surrogate objective in HPO can cause an unexpected failure, merely pursuing hyperparameters whose generalization performance is greatly overestimated. We then propose simple and computationally efficient corrections to the typical HPO procedure to deal with the aforementioned issues simultaneously. Empirical investigations demonstrate the effectiveness of our proposed HPO algorithm in situations where the typical procedure fails severely.