Improving coordination in small-scale multi-agent deep reinforcement learning through memory-driven communication

Emanuele Pesce, Giovanni Montana

Machine Learning 2020 article cited by 49

Source Open-access PDF More in Software & Coding

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

Problem

Multiple agents trained centrally but executed decentrally struggle to coordinate on tasks requiring synchronisation when they only have partial observations.

Why it matters

Inter-agent coordination is essential for multi-agent tasks requiring synchronisation, per the abstract.

What the AI did

A deep deterministic policy gradient framework with a memory device lets agents learn read/write operations and an explicit communication protocol end-to-end, building a shared world representation.

Results

Superior performance versus comparison in scenarios with up to six agents; no numeric metrics stated in abstract.

Impact

Empirically shows concurrent learning of communication and policies improves coordination in small-scale multi-agent systems.

Record

Venue	Machine Learning
Published	2020
Work type	article
Authors	Emanuele Pesce, Giovanni Montana
Citations	49 (OpenAlex, at import time)
References	68
Model or method	Deep deterministic policy gradients (DDPG) with memory-driven communication
Organization	not stated in source
Primary source	https://doi.org/10.1007/s10994-019-05864-5
Open access	hybrid
Keywords	multi-agent systems, deep reinforcement learning, communication, DDPG, coordination, memory

Importance rating

3 / 10

incremental / niche

Inter-agent coordination is a genuine research capability, but demonstrated only in small-scale scenarios (up to six agents) with no application scale.

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