Coordination Between Individual Agents in Multi-Agent Reinforcement Learning

Yang Zhang, Qingyu Yang, Dou An, Chengwei Zhang

Proceedings of the AAAI Conference on Artificial Intelligence 2021 conference-paper cited by 20

Source Open-access PDF More in Software & Coding

importance 3/10 partially solved Reinforcement Learning in Robotics multi-agent AI stage: research

Problem

Existing multi-agent reinforcement learning methods only handle global- or neighborhood-level coordination, leaving coordination between individual agents unsolved in unknown multi-agent environments.

Why it matters

Learning optimal coordinated policies is crucial for agents operating together in unknown environments.

What the AI did

Proposes an agent-level coordination MARL method combining Pearson/Spearman/Kendall correlation analysis between agents with a coordinated training framework using message dropout and correlation-based rewards.

Results

Verified in four mixed cooperative-competitive environments; quantitative results not stated in source (abstract truncated).

Impact

Method demonstrated only in test environments; no real-world deployment stated in source.

Record

Venue	Proceedings of the AAAI Conference on Artificial Intelligence
Published	2021
Work type	conference-paper
Authors	Yang Zhang, Qingyu Yang, Dou An, Chengwei Zhang
Citations	20 (OpenAlex, at import time)
References	35
Model or method	not stated in source
Organization	not stated in source
Primary source	https://doi.org/10.1609/aaai.v35i13.17357
Open access	diamond
Keywords	multi-agent reinforcement learning, coordination, correlation analysis, cooperative-competitive environments

Importance rating

3 / 10

incremental / niche

Agent-level coordination in unknown multi-agent environments; verified only in four mixed cooperative-competitive test environments, no real-world application or scale stated.

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

More in Software & Coding

#	Paper	Year	Citations	Importance
1	ImageNet classification with deep convolutional neural networks Alex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton	2017	106,345	8
2	Deep Residual Learning for Image Recognition Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun	2016	221,134	8
3	Faster sorting algorithms discovered using deep reinforcement learning Daniel J. Mankowitz, Andrea Michi, Anton Zhernov, Marco Gelmi, Marco S...	2023	175	8
4	A deep learning‐based approach for software vulnerability detection using code metrics Fazli Subhan, Xiaoxue Wu, Lili Bo, Xiaobing Sun, Muhammad Rahman	2022	42	7
5	Deep Learning-Based Attack Detection and Classification in Android Devices Alfonso Gómez, Antonio Muñoz	2023	41	7