AI Tools & Models HOTFeaturedAll
arXiv cs.AIScore65

MASTARS: Multi-Agent Sequential Trajectory Augmentation with Return-Conditioned Subgoals

Summary

Offline multi-agent reinforcement learning is limited by the quality of its fixed dataset, and MASTARS augments trajectories with return-conditioned subgoals to improve learning.

Why this matters

Covers Agent · Relevant to this topic

Open original ↗

Source: arXiv cs.AI · arxiv.org

Topics

Related items