AI Paper Atlas

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

Learning Efficient Multi-agent Cooperative Visual Exploration

Chao Yu, Xinyi Yang, Jiaxuan Gao, Huazhong Yang, Yu Wang, Yi Wu

ECCV 2022 · Multi-agent systems and games › Cooperative multi-agent RL · approach: MARL benchmarks and grouping · 34 citations (OpenAlex)

Read the paper

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
add policy distillation over per-scene teacher policies (ablation)MAANS · Habitat Gibson train scenes, N=2 agentsSteps to 90% coverage: 180.25 -> 158.55Improved
add policy distillation over per-scene teacher policies (ablation)MAANS · Habitat Gibson test scenes, N=2 agentsSteps to 90% coverage: 159.53 -> 144.16Improved
use MAANS instead of best planning baseline on train scenesRRT · Habitat Gibson train scenes, N=2 agentsSteps to 90% coverage: 199.59 -> 158.55Improved
use MAANS instead of best planning baseline on test scenesVoronoi · Habitat Gibson test scenes, N=2 agentsSteps to 90% coverage: 156.68 -> 144.16Improved
use MAANS instead of Voronoi for mutual overlap on test scenesVoronoi · Habitat Gibson test scenes, N=2 agentsMutual Overlap: 0.51 -> 0.54Hurt
replace multi-agent planners APF and WMA-RRT with single-agent RRTAPF/WMA-RRT · Habitat Gibson train scenes, N=2 agentsSteps to 90% coverage: 268.20 -> 199.59No clear effect
zero-shot transfer MAANS to N=3 on train scenes versus VoronoiVoronoi · Habitat Gibson train, N=3Steps to 90% coverage: 155.13 -> 127.88Improved
zero-shot transfer MAANS to N=4 on train scenes versus RRTRRT · Habitat Gibson train, N=4Steps to 90% coverage: 140.57 -> 114.75Improved
zero-shot transfer MAANS to N=4 on test scenes versus VoronoiVoronoi · Habitat Gibson test, N=4Steps to 90% coverage: 101.90 -> 109.07Hurt
team size increases mid-episode (2 to 3) on train scenesRRT · Habitat Gibson train, 2=>3Steps to 90% coverage: 185.94 -> 148.82Improved
team size decreases mid-episode (3 to 2) on train scenesRRT · Habitat Gibson train, 3=>2Steps to 90% coverage: 187.93 -> 145.14Improved
add Spatial-TeamFormer instead of average pooling (ablation)MSP · Gibson Colebrook and DryvilleStepsImproved

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