AI Paper Atlas

Research map / Multi-agent systems and games / Human-AI and goal-conditioned

The Neural MMO Platform for Massively Multiagent Research

Joseph Suarez, Yilun Du, Clare Zhu, Igor Mordatch, Phillip Isola

NeurIPS 2021 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: World-model-based agents · 1 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
train on 1 map vs 32 procedurally generated maps (ablation)PPO LSTM · unseen mapsaverage lifetime: 52.73 -> 61.56Improved
32 vs 256 training maps (ablation)PPO LSTM · unseen mapsaverage lifetime: 61.56 -> 61.91No clear effect
16384 training maps vs 32 (ablation)PPO LSTM · unseen mapsaverage lifetime: 61.56 -> 53.25No clear effect
larger training population cap (4 -> 256) (ablation)PPO LSTM · SmallMaps resource+progressionlifetime: 89.26 -> 227.36Improved
larger training population cap (4 -> 32) (ablation)PPO LSTM · SmallMaps resource+progressionexplore: 7.66 -> 11.41Improved
remove exploration routine from scripted forage bot (ablation)Scripted Forage · SmallMapsachievement: 7.56 -> 4.34Hurt
remove exploration routine from scripted combat bot (ablation)Scripted Combat · SmallMapsachievement: 3.45 -> 2.72Hurt
add combat heuristic to scripted forage botScripted bots · SmallMapslifetime: 252.38 -> 76.52Hurt
transfer SmallMaps policy to LargeMaps zero-shotNeural MMO policy · LargeMapslifetime: 57.12 -> 73.48Mixed
evaluate LargeMaps policy zero-shot on SmallMapsNeural MMO policy · SmallMapslifetime: 51.86 -> 35.24Hurt
scale maps from SmallMaps to LargeMaps for neural forageNeural Forage · Neural MMOlifetime: 132.14 -> 356.13Improved
increase test-time population cap for agents trained in small populationsPPO LSTM · Neural MMOlifetimeHurt

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