DRATS derives a minimax objective from a feasibility formulation of MTRL to adaptively sample tasks with the largest return gaps, leading to better worst-task performance on MetaWorld benchmarks.
Sample Complexity of Multi-task Reinforcement Learning
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
abstract
Transferring knowledge across a sequence of reinforcement-learning tasks is challenging, and has a number of important applications. Though there is encouraging empirical evidence that transfer can improve performance in subsequent reinforcement-learning tasks, there has been very little theoretical analysis. In this paper, we introduce a new multi-task algorithm for a sequence of reinforcement-learning tasks when each task is sampled independently from (an unknown) distribution over a finite set of Markov decision processes whose parameters are initially unknown. For this setting, we prove under certain assumptions that the per-task sample complexity of exploration is reduced significantly due to transfer compared to standard single-task algorithms. Our multi-task algorithm also has the desired characteristic that it is guaranteed not to exhibit negative transfer: in the worst case its per-task sample complexity is comparable to the corresponding single-task algorithm.
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
A low-rank matrix estimation method in a reward-free RL framework learns shared representations across linear MDPs and yields near-optimal policies with characterized regret bounds under relaxed feature assumptions.
citing papers explorer
-
Distributionally Robust Multi-Task Reinforcement Learning via Adaptive Task Sampling
DRATS derives a minimax objective from a feasibility formulation of MTRL to adaptively sample tasks with the largest return gaps, leading to better worst-task performance on MetaWorld benchmarks.
-
Provable Multi-Task Reinforcement Learning: A Representation Learning Framework with Low Rank Rewards
A low-rank matrix estimation method in a reward-free RL framework learns shared representations across linear MDPs and yields near-optimal policies with characterized regret bounds under relaxed feature assumptions.