analiza aiqbench
This research introduces ACC-MARL, a framework that uses automata to decompose complex cooperative objectives into manageable sub-tasks for multi-agent systems. By proving optimality and enabling task-conditioned policies, the approach overcomes sample inefficiency and single-task limitations inherent in existing decentralized execution methods. The work demonstrates emergent, multi-step coordination capabilities, offering a scalable solution for dynamic team-level objectives without requiring retraining for new tasks.
Automata-Conditioned Cooperative Multi-Agent Reinforcement Learning
arXiv:2511.02304v2 Announce Type: replace-cross Abstract: We study learning multi-task, multi-agent policies for cooperative, temporal objectives, under centralized training, decentralized execution. In this setting, using automata to represent tasks assigned to agents enables breaking down a team-level objective into simpler, smaller sub-tasks. However, existing approaches remain sample-inefficient and are limited to the single-task case, requiring retraining policies for each new task. In this work, we present Automata-Conditioned Cooperative Multi-Agent Reinforcement Learning (ACC-MARL), a framework for learning task-conditioned, decentralized team policies. We identify challenges to the feasibility of ACC-MARL, propose solutions, and prove that our approach is optimal. We further show that learned value functions can be used to assign tasks optimally at test time. Experiments demonstrate emergent task-aware, multi-step coordination among agents, such as pressing a button to unlock a door, holding the door, and short-circuiting tasks.
Abstract:We study learning multi-task, multi-agent policies for cooperative, temporal objectives, under centralized training, decentralized execution. In this setting, using automata to represent tasks assigned to agents enables breaking down a team-level objective into simpler, smaller sub-tasks. However, existing approaches remain sample-inefficient and are limited to the single-task case, requiring retraining policies for each new task. In this work, we present Automata-Conditioned Cooperative Multi-Agent Reinforcement Learning (ACC-MARL), a framework for learning task-conditioned, decentralized team policies. We identify challenges to the feasibility of ACC-MARL, propose solutions, and prove that our approach is optimal. We further show that learned value functions can be used to assign tasks optimally at test time. Experiments demonstrate emergent task-aware, multi-step coordination among agents, such as pressing a button to unlock a door, holding the door, and short-circuiting tasks.
Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Formal Languages and Automata Theory (cs.FL); Machine Learning (cs.LG)
arXiv:2511.02304 [cs.MA]
(or arXiv:2511.02304v2 [cs.MA] for this version)
Text integral sub licența deschisă a publicației · 158 cuvinte