English

Accommodating Picky Customers: Regret Bound and Exploration Complexity for Multi-Objective Reinforcement Learning

Machine Learning 2021-10-29 v3 Machine Learning

Abstract

In this paper we consider multi-objective reinforcement learning where the objectives are balanced using preferences. In practice, the preferences are often given in an adversarial manner, e.g., customers can be picky in many applications. We formalize this problem as an episodic learning problem on a Markov decision process, where transitions are unknown and a reward function is the inner product of a preference vector with pre-specified multi-objective reward functions. We consider two settings. In the online setting, the agent receives a (adversarial) preference every episode and proposes policies to interact with the environment. We provide a model-based algorithm that achieves a nearly minimax optimal regret bound O~(min{d,S}H2SAK)\widetilde{\mathcal{O}}\bigl(\sqrt{\min\{d,S\}\cdot H^2 SAK}\bigr), where dd is the number of objectives, SS is the number of states, AA is the number of actions, HH is the length of the horizon, and KK is the number of episodes. Furthermore, we consider preference-free exploration, i.e., the agent first interacts with the environment without specifying any preference and then is able to accommodate arbitrary preference vector up to ϵ\epsilon error. Our proposed algorithm is provably efficient with a nearly optimal trajectory complexity O~(min{d,S}H3SA/ϵ2)\widetilde{\mathcal{O}}\bigl({\min\{d,S\}\cdot H^3 SA}/{\epsilon^2}\bigr). This result partly resolves an open problem raised by \citet{jin2020reward}.

Keywords

Cite

@article{arxiv.2011.13034,
  title  = {Accommodating Picky Customers: Regret Bound and Exploration Complexity for Multi-Objective Reinforcement Learning},
  author = {Jingfeng Wu and Vladimir Braverman and Lin F. Yang},
  journal= {arXiv preprint arXiv:2011.13034},
  year   = {2021}
}

Comments

NeurIPS 2021 Camera Ready Version

R2 v1 2026-06-23T20:31:02.821Z