HHC: Hierarchical Hypergraph Communication for Multi-Agent Systems

Qifan Liang, Chenlong Li, Feiyu Wang, Jing Fu, Yixiang Shan, Lu Guo, Wei Liu, Lichang Song, Ting Long
Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence, PMLR 337:3802-3818, 2026.

Abstract

Cooperative multi-agent reinforcement learning ({MARL}) faces significant coordination challenges due to partial observability. Although communication can mitigate these issues, traditional methods often overlook the fact that agents can simultaneously belong to multiple collaborative groups, playing diverse roles. This limitation hinders the effective integration of tactical and strategic information. To address this, we propose Hierarchical Hypergraph Communication (HHC). In HHC, agents are modeled via a tactical-level hypergraph that supports overlapping multi-group memberships, facilitating the extraction of fine-grained tactical features. Simultaneously, these tactical hyperedges serve as virtual nodes to integrate and communicate strategic intents within a strategic-level hypergraph. This dual-layered architecture empowers agents to align individual actions with global strategies through a hierarchical communication process, thereby enhancing decision-making capabilities. Extensive experiments demonstrate the effectiveness of HHC in complex coordination tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v337-liang26a, title = {HHC: Hierarchical Hypergraph Communication for Multi-Agent Systems}, author = {Liang, Qifan and Li, Chenlong and Wang, Feiyu and Fu, Jing and Shan, Yixiang and Guo, Lu and Liu, Wei and Song, Lichang and Long, Ting}, booktitle = {Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence}, pages = {3802--3818}, year = {2026}, editor = {Perković, Emilija and Malinsky, Daniel}, volume = {337}, series = {Proceedings of Machine Learning Research}, month = {17--21 Aug}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v337/main/assets/liang26a/liang26a.pdf}, url = {https://proceedings.mlr.press/v337/liang26a.html}, abstract = {Cooperative multi-agent reinforcement learning ({MARL}) faces significant coordination challenges due to partial observability. Although communication can mitigate these issues, traditional methods often overlook the fact that agents can simultaneously belong to multiple collaborative groups, playing diverse roles. This limitation hinders the effective integration of tactical and strategic information. To address this, we propose Hierarchical Hypergraph Communication (HHC). In HHC, agents are modeled via a tactical-level hypergraph that supports overlapping multi-group memberships, facilitating the extraction of fine-grained tactical features. Simultaneously, these tactical hyperedges serve as virtual nodes to integrate and communicate strategic intents within a strategic-level hypergraph. This dual-layered architecture empowers agents to align individual actions with global strategies through a hierarchical communication process, thereby enhancing decision-making capabilities. Extensive experiments demonstrate the effectiveness of HHC in complex coordination tasks.} }
Endnote
%0 Conference Paper %T HHC: Hierarchical Hypergraph Communication for Multi-Agent Systems %A Qifan Liang %A Chenlong Li %A Feiyu Wang %A Jing Fu %A Yixiang Shan %A Lu Guo %A Wei Liu %A Lichang Song %A Ting Long %B Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2026 %E Emilija Perković %E Daniel Malinsky %F pmlr-v337-liang26a %I PMLR %P 3802--3818 %U https://proceedings.mlr.press/v337/liang26a.html %V 337 %X Cooperative multi-agent reinforcement learning ({MARL}) faces significant coordination challenges due to partial observability. Although communication can mitigate these issues, traditional methods often overlook the fact that agents can simultaneously belong to multiple collaborative groups, playing diverse roles. This limitation hinders the effective integration of tactical and strategic information. To address this, we propose Hierarchical Hypergraph Communication (HHC). In HHC, agents are modeled via a tactical-level hypergraph that supports overlapping multi-group memberships, facilitating the extraction of fine-grained tactical features. Simultaneously, these tactical hyperedges serve as virtual nodes to integrate and communicate strategic intents within a strategic-level hypergraph. This dual-layered architecture empowers agents to align individual actions with global strategies through a hierarchical communication process, thereby enhancing decision-making capabilities. Extensive experiments demonstrate the effectiveness of HHC in complex coordination tasks.
APA
Liang, Q., Li, C., Wang, F., Fu, J., Shan, Y., Guo, L., Liu, W., Song, L. & Long, T.. (2026). HHC: Hierarchical Hypergraph Communication for Multi-Agent Systems. Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 337:3802-3818 Available from https://proceedings.mlr.press/v337/liang26a.html.

Related Material