Provably Efficient Algorithms for Multi-Objective Competitive RL
This work provides the first provably efficient algorithms for vector-valued Markov games, which is a foundational problem for multi-objective competitive reinforcement learning.
This paper addresses multi-objective reinforcement learning in competitive settings, where an agent's reward is a vector and performance is measured by the distance of its average return vector to a target set. The authors developed statistically and computationally efficient algorithms that approach the target set, extending Blackwell's approachability theorem to tabular RL and achieving near-optimal theoretical guarantees.
We study multi-objective reinforcement learning (RL) where an agent's reward is represented as a vector. In settings where an agent competes against opponents, its performance is measured by the distance of its average return vector to a target set. We develop statistically and computationally efficient algorithms to approach the associated target set. Our results extend Blackwell's approachability theorem (Blackwell, 1956) to tabular RL, where strategic exploration becomes essential. The algorithms presented are adaptive; their guarantees hold even without Blackwell's approachability condition. If the opponents use fixed policies, we give an improved rate of approaching the target set while also tackling the more ambitious goal of simultaneously minimizing a scalar cost function. We discuss our analysis for this special case by relating our results to previous works on constrained RL. To our knowledge, this work provides the first provably efficient algorithms for vector-valued Markov games and our theoretical guarantees are near-optimal.