aiqbench analysis
This paper presents MARLIN, a hybrid framework that integrates large language models into multi-agent reinforcement learning to guide robot negotiation and planning during early training stages. By dynamically switching between LLM-based high-level planning and standard RL policy learning, the approach improves safety and performance in simulated and physical robotic systems without compromising final outcomes. This work advances the intersection of generative AI and robotics control, offering a practical method for safer multi-robot system deployment.
MARLIN: Multi-Agent Reinforcement Learning Guided by Language-Based Inter-Robot Negotiation
arXiv:2410.14383v5 Announce Type: replace Abstract: Multi-agent reinforcement learning is a key method for training multi-robot systems. Through rewarding or punishing robots over a series of episodes according to their performance, they can be trained and then deployed in the real world. However, poorly trained policies can lead to unsafe behaviour during early training stages. We introduce Multi-Agent Reinforcement Learning guided by language-based Inter-robot Negotiation (MARLIN), a hybrid framework in which large language models provide high-level planning before the reinforcement learning policy has learned effective behaviours. Robots use language models to negotiate actions and generate plans that guide policy learning. The system dynamically switches between reinforcement learning and language-model-based negotiation during training, enabling more effective exploration. MARLIN is evaluated using both simulated and physical robots with local and remote language models. Results show that, compared to standard multi-agent reinforcement learning, the hybrid approach achieves higher performance in early training with limited impact on the final performance. The code is available at https://github.com/SooratiLab/MARLIN.
Abstract:Multi-agent reinforcement learning is a key method for training multi-robot systems. Through rewarding or punishing robots over a series of episodes according to their performance, they can be trained and then deployed in the real world. However, poorly trained policies can lead to unsafe behaviour during early training stages. We introduce Multi-Agent Reinforcement Learning guided by language-based Inter-robot Negotiation (MARLIN), a hybrid framework in which large language models provide high-level planning before the reinforcement learning policy has learned effective behaviours. Robots use language models to negotiate actions and generate plans that guide policy learning. The system dynamically switches between reinforcement learning and language-model-based negotiation during training, enabling more effective exploration. MARLIN is evaluated using both simulated and physical robots with local and remote language models. Results show that, compared to standard multi-agent reinforcement learning, the hybrid approach achieves higher performance in early training with limited impact on the final performance. The code is available at this https URL.
6 pages, 7 figures, 1 table. Accepted at ROBIO2026
Robotics (cs.RO)
arXiv:2410.14383 [cs.RO]
(or arXiv:2410.14383v5 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2410.14383 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 258 words