EUVIMEDEUVIMED entdecken
Crossref

Offline Multi-Objective Bandits: From Logged Data to Pareto-Optimal Policies

Ji Cheng, Song Lai, Shunyu Yao, Bo Xue

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (43) · 36636-36644

DOI: 10.1609/aaai.v40i43.40987 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Ji Cheng und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Offline policy learning from logged data is a critical paradigm for enabling effective decision-making without costly online exploration. However, its application has been largely confined to single-objective problems, a stark contrast to real-world scenarios where decision-making inherently involves navigating multiple, often conflicting, objectives. This paper introduces a comprehensive framework for Offline Multi-Objective Bandits (OffMOB), providing a principled solution to the fundamental challenge of learning Pareto-optimal policies from a static dataset. Our core contribution is a novel algorithm that uniquely integrates the pessimism principle with multi-objective optimization to safely learn from off-policy data. Crucially, our approach transcends the primary limitation of scalarization techniques, which are restricted to finding a single policy for a pre-defined preference. Instead, OffMOB directly approximates the entire Pareto front, learning a single, flexible policy model capable of generating an optimal action for any desired trade-off. To rigorously evaluate performance, we introduce the Tchebycheff sub-optimality metric and establish the first finite-sample generalization bounds for this problem class, proving that our algorithm converges to the true Pareto front under practical data coverage assumptions. Extensive experiments on complex benchmarks demonstrate that OffMOB significantly outperforms existing methods, identifying the complete set of optimal trade-offs where naive extensions fail.

Zitieren

APA

Cheng, J., Lai, S., Yao, S., & Xue, B. (2026). Offline Multi-Objective Bandits: From Logged Data to Pareto-Optimal Policies. Proceedings of the AAAI Conference on Artificial Intelligence, 40(43), 36636-36644. https://doi.org/10.1609/aaai.v40i43.40987

Vancouver

Cheng J, Lai S, Yao S, Xue B. Offline Multi-Objective Bandits: From Logged Data to Pareto-Optimal Policies. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(43):36636-36644. doi:10.1609/aaai.v40i43.40987

BibTeX
@article{Cheng2026,
  author = {Cheng, Ji and Lai, Song and Yao, Shunyu and Xue, Bo},
  title = {Offline Multi-Objective Bandits: From Logged Data to Pareto-Optimal Policies},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {43},
  pages = {36636--36644},
  doi = {10.1609/aaai.v40i43.40987}
}
Kostenlos starten →