HAVEN: Hierarchical Cooperative Multi-Agent Reinforcement Learning with Dual Coordination Mechanism

Zhiwei Xu, Yunpeng Bai, Bin Zhang, Dapeng Li, Guoliang Fan

Proceedings of the AAAI Conference on Artificial Intelligence 2023 conference-paper cited by 25

Source Open-access PDF More in Robotics & Autonomy

importance 4/10 ai assisted Reinforcement Learning in Robotics multi-agent RL research stage: research

Problem

Instability arises when policies of multiple levels and agents are optimized concurrently in fully cooperative multi-agent tasks.

Why it matters

Reliable multi-agent coordination is needed for cooperative real-world systems.

What the AI did

HAVEN, a hierarchical value decomposition framework, introduces a dual coordination mechanism (inter-level and inter-agent) via reward function design in a two-level hierarchy, requiring no domain knowledge or pre-training.

Results

Achieves desirable results on Dec-POMDP domains and outperforms popular multi-agent hierarchical RL algorithms (no numbers stated).

Impact

Evidence supports a benchmark advantage over prior algorithms in a top-tier peer-reviewed venue.

Record

Venue	Proceedings of the AAAI Conference on Artificial Intelligence
Published	2023
Work type	conference-paper
Authors	Zhiwei Xu, Yunpeng Bai, Bin Zhang, Dapeng Li, Guoliang Fan
Citations	25 (OpenAlex, at import time)
References	68
Model or method	HAVEN (hierarchical value decomposition with dual coordination mechanism)
Organization	not stated in source
Primary source	https://doi.org/10.1609/aaai.v37i10.26386
Open access	diamond
Keywords	multi-agent reinforcement learning, hierarchical RL, value decomposition, cooperation, Dec-POMDP

Importance rating

4 / 10

real but narrower advance

Addresses instability in cooperative multi-agent RL, a foundational methods problem, but abstract only shows benchmark gains, no real-world scale.

Basis: inferred. Judged from the source text, which gives no explicit scale. Stated scale: not stated in source. Method and limits.

More in Robotics & Autonomy

#	Paper	Year	Citations	Importance
1	Autonomous robotic laparoscopic surgery for intestinal anastomosis Hamed Saeidi, Justin David Opfermann, Michael Kam, Shuwen Wei, Simon L...	2022	384	7
2	Solving Rubik's Cube with a Robot Hand OpenAI, Ilge Akkaya, Marcin Andrychowicz, Maciek Chociej, Mateusz Litw...	2019	569	6
3	PaLM-E: An Embodied Multimodal Language Model Danny Driess, Fei Xia, Mehdi S. M. Sajjadi, Corey Lynch, Aakanksha Cho...	2023	346	6
4	Assistive Navigation Using Deep Reinforcement Learning Guiding Robot With UWB/Voice Beacons and Semantic Feedbacks for Blind and Visually Impaired People Chen-Lung Lu, Ziyan Liu, Jui-Te Huang, Ching-I Huang, Bo-Hui Wang, Yi...	2021	48	6
5	Towards Deep Radar Perception for Autonomous Driving: Datasets, Methods, and Challenges Yi Zhou, Lulu Liu, Haocheng Zhao, Miguel López‐Benítez, Limin Yu, Yuta...	2022	154	6