MASTARS: Multi-Agent Sequential Trajectory Augmentation with Return-Conditioned Subgoals
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
The Flow has not summarised this story yet — read it at arXiv AI.
arXiv:2307. 01472v2 Announce Type: replace Abstract: We present a novel Diffusion Offline Multi-agent Model (DOM2) for offline Multi-Agent Reinforcement Learning (MARL).
The paper introduces Coordinated Multi-Agent Diffusion Steering (CMDS), a framework that reuses independently trained diffusion models as generative primitives and learns a control policy to coordinate their reverse processes. By framing coordination as a stochastic optimal control problem, CMDS balances an assembly-level reward with adherence to pretrained dynamics, enabling the learned control to generalise across new tasks. Experiments demonstrate CMDS’s ability to recover target distributions, satisfy spatial constraints, and separate sources from degraded mixtures in applications such as maze navigation, robot planning, and text-conditioned human motion.
arXiv:2602. 18291v2 Announce Type: replace Abstract: Online Multi-Agent Reinforcement Learning (MARL) is a prominent framework for efficient agent coordination.
arXiv:2506. 01568v4 Announce Type: replace Abstract: Being able to solve a task in diverse ways makes agents more robust to task variations and less prone to local optima.
arXiv:2606. 25073v1 Announce Type: new Abstract: In cooperative multi-agent reinforcement learning (MARL), from a deployment perspective, it is challenging and expensive to train agents from scratch for each new environment or task.
The paper investigates zero‑shot task generalisation in offline multi‑agent reinforcement learning by extending sequence‑modeling architectures to support multi‑task observation and action spaces and variable agent counts. It finds that increasing task diversity, rather than merely enlarging the dataset, is the key driver for robust zero‑shot transfer. Experiments on four challenging environments show a 3.2× mean improvement on held‑out tasks compared to single‑task models and outperform strong behaviour‑cloning baselines.