Work

Optimal Baseline Corrections for Off-Policy Contextual Bandits

Year: 2024

Type: conference-paper

Abstract: The off-policy learning paradigm allows for recommender systems and general ranking applications to be framed as decision-making problems, where we aim to learn decision policies that optimize an unbiased offline estimate of an online reward metric. With unbiasedness comes potentially high variance, and prevalent methods exist to reduce estimation variance. These methods typically make use of control variates, either additive (i.e., baseline corrections or doubly robust methods) or multiplicative (i.e., self-normalisation).

Source: ACM Conference on Recommender Systems (RecSys)

DOI: 10.1145/3640457.3688105

Authors: Shashank Gupta, Olivier Jeunen, Harrie Oosterhuis, Maarten de Rijke

Institutions: University of Amsterdam, Radboud University Nijmegen

Language: English

Topic: Advanced Bandit Algorithms Research

Subfield: Management Science and Operations Research

Field: Decision Sciences

Domain: Social Sciences

Open Access status: hybrid

Funders: European Commission, Nederlandse Organisatie voor Wetenschappelijk Onderzoek, HORIZON EUROPE Framework Programme

Awards: Fairness and Intersectional Non-Discrimination in Human Recommendation, Hybrid Intelligence (HI): augmenting human intellect, Low Resource Chat-based Conversational Intelligence, Trustworthy AI-based Systems for Sustainable Growth