Pith. sign in

REVIEW 2 cited by

Bayesian Off-Policy Evaluation and Learning for Large Action Spaces

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2402.14664 v2 pith:TSIQIDYJ submitted 2024-02-22 cs.LG cs.AIstat.ML

Bayesian Off-Policy Evaluation and Learning for Large Action Spaces

classification cs.LG cs.AIstat.ML
keywords bayesianactioncorrelationsevaluationframeworkintroducelargelearning
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

In interactive systems, actions are often correlated, presenting an opportunity for more sample-efficient off-policy evaluation (OPE) and learning (OPL) in large action spaces. We introduce a unified Bayesian framework to capture these correlations through structured and informative priors. In this framework, we propose sDM, a generic Bayesian approach for OPE and OPL, grounded in both algorithmic and theoretical foundations. Notably, sDM leverages action correlations without compromising computational efficiency. Moreover, inspired by online Bayesian bandits, we introduce Bayesian metrics that assess the average performance of algorithms across multiple problem instances, deviating from the conventional worst-case assessments. We analyze sDM in OPE and OPL, highlighting the benefits of leveraging action correlations. Empirical evidence showcases the strong performance of sDM.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Offline Contextual Bandit with Counterfactual Sample Identification

    cs.LG 2025-09 conditional novelty 5.0

    CSI converts offline contextual bandit learning into a binary classification problem by comparing the logged action against a counterfactual action sampled from the logging policy, and the argmax of the resulting clas...

  2. Off-Policy Learning in Large Action Spaces: Optimization Matters More Than Estimation

    stat.ML 2025-09 conditional novelty 4.0

    Reward-weighted log-likelihood objectives outperform complex off-policy estimators in large action spaces because their optimization landscapes are much easier to navigate.