2024

Effective Off-Policy Evaluation and Learning in Contextual Combinatorial Bandits
We explore off-policy evaluation and learning (OPE/L) in contextual combinatorial bandits (CCB), where a policy selects a subset in the …
Tatsuhiro Shimizu, Koichi Tanaka, Ren Kishimoto, Haruka Kiyohara, Masahiro Nomura, Yuta Saito
Cite
Hyperparameter Optimization Can Even be Harmful in Off-Policy Learning and How to Deal with It
There has been a growing interest in off-policy evaluation in the literature such as recommender systems and personalized medicine. We …
Yuta Saito, Masahiro Nomura
Cite Proceedings
Long-term Off-Policy Evaluation and Learning
Short- and long-term outcomes of an algorithm often differ, with damaging downstream effects. A known example is a click-bait …
Yuta Saito, Himan Abdollahpouri, Jesse Anderton, Ben Carterette, Mounia Lalmas
Cite Code arXiv Proceedings
Off-Policy Evaluation of Slate Bandit Policies via Optimizing Abstraction
We study off-policy evaluation (OPE) in slate contextual bandits where a policy selects multi-dimensional actions known as slates. This …
Haruka Kiyohara, Masahiro Nomura, Yuta Saito
Cite Code arXiv Proceedings
Scalable and Provably Fair Exposure Control for Large-Scale Recommender Systems
Typical recommendation and ranking methods aim to optimize the satisfaction of users, but they are often oblivious to their impact on …
Riku Togashi, Kenshi Abe, Yuta Saito
Cite Video arXiv Proceedings
Towards Assessing and Benchmarking Risk-Return Tradeoff of Off-Policy Evaluation
Off-Policy Evaluation (OPE) aims to assess the effectiveness of counterfactual policies using only offline logged data and is often …
Haruka Kiyohara, Ren Kishimoto, Kosuke Kawakami, Ken Kobayashi, Kazuhide Nakata, Yuta Saito
Cite Code arXiv