EUVIMEDEUVIMED entdecken
Crossref

Centralized Training with Hybrid Execution in Multi-Agent Reinforcement Learning via Predictive Observation Imputation (Abstract Reprint)

Pedro P. Santos, Diogo S. Carvalho, Miguel Vasco, Alberto Sardinha, Pedro A. Santos, Ana Paiva, Francisco S. Melo

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (47) · 39894-39894

DOI: 10.1609/aaai.v40i47.41409 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Pedro P. Santos und 6 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We study hybrid execution in multi-agent reinforcement learning (MARL), a paradigm where agents aim to complete cooperative tasks with arbitrary communication levels at execution time by taking advantage of information-sharing among the agents. Under hybrid execution, the communication level can range from a setting in which no communication is allowed between agents (fully decentralized), to a setting featuring full communication (fully centralized), but the agents do not know beforehand which communication level they will encounter at execution time. We contribute MARO, an approach that makes use of an auto-regressive predictive model, trained in a centralized manner, to estimate missing agents' observations at execution time. We evaluate MARO on standard scenarios and extensions of previous benchmarks tailored to emphasize the impact of partial observability in MARL. Experimental results show that our method consistently outperforms relevant baselines, allowing agents to act with faulty communication while successfully exploiting shared information.

Zitieren

APA

Santos, P. P., Carvalho, D. S., Vasco, M., Sardinha, A., Santos, P. A., Paiva, A., & Melo, F. S. (2026). Centralized Training with Hybrid Execution in Multi-Agent Reinforcement Learning via Predictive Observation Imputation (Abstract Reprint). Proceedings of the AAAI Conference on Artificial Intelligence, 40(47), 39894-39894. https://doi.org/10.1609/aaai.v40i47.41409

Vancouver

Santos PP, Carvalho DS, Vasco M, Sardinha A, Santos PA, Paiva A, et al. Centralized Training with Hybrid Execution in Multi-Agent Reinforcement Learning via Predictive Observation Imputation (Abstract Reprint). Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(47):39894-39894. doi:10.1609/aaai.v40i47.41409

BibTeX
@article{Santos2026,
  author = {Santos, Pedro P. and Carvalho, Diogo S. and Vasco, Miguel and Sardinha, Alberto and Santos, Pedro A. and Paiva, Ana and Melo, Francisco S.},
  title = {Centralized Training with Hybrid Execution in Multi-Agent Reinforcement Learning via Predictive Observation Imputation (Abstract Reprint)},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {47},
  pages = {39894--39894},
  doi = {10.1609/aaai.v40i47.41409}
}
Kostenlos starten →