analiza aiqbench
This paper introduces MA-FPPO, a novel framework that combines flow-matching pretraining with online fine-tuning to enhance multi-agent coordination in unseen environments. By leveraging shared team advantages for continuous and discrete action spaces, the method delivers significant performance gains over both offline baselines and purely online learning approaches. The work addresses a critical gap in robust cooperative behavior, offering a practical path forward for deploying multi-agent systems in dynamic real-world scenarios.
MA-FPPO: Multi-Agent Flow-Pretrained Policy Optimization
arXiv:2609.32594v1 Announce Type: new Abstract: Multi-agent flow policies learn cooperative behavior from fixed offline datasets, but often struggle to complete tasks in situations not covered by the offline data. In these situations, agents must both adapt to changes in the environment and coordinate with one another, yet action patterns learned offline are often insufficient for effective adaptation and coordination. To address this problem, we propose Multi-Agent Flow-Pretrained Policy Optimization (MA-FPPO), which uses online fine-tuning to improve the cooperative behavior of models pretrained with flow matching through new interactions with the environment. Building on the behavior learned during pretraining, we construct policies with explicit action likelihoods for discrete and continuous action spaces. We then update the pretrained model using shared team advantages to further improve coordination based on team performance. Our method achieves, on average, relative gains of 52.8% over the strongest listed offline baselines across 30 settings and 29.8% over purely online learning across 38 comparisons with matched online budgets and evaluation protocols.
Abstract:Multi-agent flow policies learn cooperative behavior from fixed offline datasets, but often struggle to complete tasks in situations not covered by the offline data. In these situations, agents must both adapt to changes in the environment and coordinate with one another, yet action patterns learned offline are often insufficient for effective adaptation and coordination. To address this problem, we propose Multi-Agent Flow-Pretrained Policy Optimization (MA-FPPO), which uses online fine-tuning to improve the cooperative behavior of models pretrained with flow matching through new interactions with the environment. Building on the behavior learned during pretraining, we construct policies with explicit action likelihoods for discrete and continuous action spaces. We then update the pretrained model using shared team advantages to further improve coordination based on team performance. Our method achieves, on average, relative gains of 52.8% over the strongest listed offline baselines across 30 settings and 29.8% over purely online learning across 38 comparisons with matched online budgets and evaluation protocols.
40 pages, 11 figures. Project page: this https URL
Artificial Intelligence (cs.AI)
arXiv:2609.32594 [cs.AI]
(or arXiv:2609.32594v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.32594 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
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.
Text integral sub licența deschisă a publicației · 261 cuvinte
Subiecte conexe
[AINews] OpenAI reports Navier-Stokes singularity find in 88 hours using Astra-next, roughly 10,000 agents and 130B tokens (>$40M), a contender for second ever Millennium Prize awarded
Latent Space · 9 sept. 2026
Links #5: 2026/07
LessWrong · 31 iul. 2026
Reflection’s Beam model hits 501 billion parameters
Quantum Zeitgeist · 5 oct. 2026
Flow generation through natural language: An agentic modeling approach
Shopify Engineering Blog · 22 apr. 2026