AI Paper Atlas

Research map / Multi-agent systems and games / Cooperative multi-agent RL

IMP-MARL: a Suite of Environments for Large-scale Infrastructure Management Planning via MARL

Pascal Leroy, Pablo G. Morato, Jonathan Pisane, Athanasios Kolios, Damien Ernst

NeurIPS 2023 · Multi-agent systems and games › Cooperative multi-agent RL · approach: Cooperative MARL algorithms · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use MARL methods instead of expert heuristic policiesCTDE MARL methods · IMP-MARL environmentsnormalised discounted reward vs heuristicImproved
use CTDE methods (QMIX, QVMIX, QPLEX) instead of independent Q-learningIQL · IMP-MARL environmentsnormalised discounted reward vs heuristicImproved
use CTDE methods instead of centralised DQNDQN · IMP-MARL environmentsnormalised discounted reward vs heuristicImproved
increase the number of agents for centralised DQNDQN · IMP-MARL environmentsnormalised discounted reward vs heuristicHurt
add correlation between components in k-out-of-n systemsMARL methods · correlated k-out-of-nnormalised discounted reward vs heuristicHurt
add a global campaign cost with fewer than 50 agentsMARL methods · campaign cost environmentsnormalised discounted reward vs heuristicHurt
add a global campaign cost with more than 50 agentsMARL methods · campaign cost environmentsnormalised discounted reward vs heuristicNo clear effect
increase number of agents in offshore wind farm environmentsMARL methods · offshore wind farmnormalised discounted reward vs heuristicHurt
increase the number of agents under campaign costMARL methods · campaign cost environmentsnormalised discounted reward vs heuristicImproved
use COMA with more agentsCOMA · IMP-MARL environmentsnormalised discounted reward vs heuristicHurt
use FACMAC with over 50 agentsFACMAC · IMP environments over 50 agentsnormalised discounted reward vs heuristicImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy