AI Paper Atlas

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

Distributed Inverse Constrained Reinforcement Learning for Multi-agent Systems

Shicheng Liu, Minghui Zhu

NeurIPS 2022 · Multi-agent systems and games › Cooperative multi-agent RL · approach: Decentralized actor-critic MARL · 3 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
distributed learners (D-ICRL) vs centralized C-ICRLD-ICRL · Synthetic grid world CMGcumulative reward (CR): 1890.67 ± 135.22 vs 1884.74 ± 158.90No clear effect
D-ICRL vs ME-greedyD-ICRL · Synthetic grid world CMGcumulative reward (CR): 1162.24 -> 1890.67Improved
D-ICRL vs ME-greedyD-ICRL · Synthetic grid world CMGFNR: 0.24 -> 0.18Improved
D-ICRL vs ME-greedyD-ICRL · Synthetic grid world CMGFPR: 0.18 -> 0.00Improved
D-ICRL vs MCE-greedyD-ICRL · Synthetic grid world CMGcumulative reward (CR): 1776.84 -> 1890.67Improved
D-ICRL vs MCE-greedyD-ICRL · Synthetic grid world CMGFNR: 0.12 -> 0.18Hurt
D-ICRL vs MCE-greedyD-ICRL · Synthetic grid world CMGCVR: 0.057 -> 0.013Improved
D-ICRL vs C-ICRLD-ICRL · Synthetic grid world CMGRFD: 0.731 -> 0.748No clear effect
D-ICRL vs M(C)E-greedyD-ICRL · Drones motion planning with obstaclessuccess rate (SR): 0.84 -> 0.97Improved
D-ICRL vs M(C)E-greedyD-ICRL · Drones motion planning with obstaclesFNR: 0.13 -> 0.00Improved
D-ICRL vs C-ICRLD-ICRL · Drones motion planning with obstaclessuccess rate (SR): 0.98 -> 0.97No clear effect

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