AI Paper Atlas

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

A Robust and Opponent-Aware League Training Method for StarCraft II

Ruozi Huang, Xipeng Wu, Hongsheng Yu, Zhong Fan, Haobo Fu, QIANG FU, Yang Wei

NeurIPS 2023 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Human-AI coordination agents · 0 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
add opponent modeling embedding and scouting reward to AlphaStar (ablation)AlphaStar MA · VC Stalker Play opponentwin rate: 22% -> 52%Improved
add opponent modeling to AlphaStar (ablation)AlphaStar MA · Void Ray push opponentwin rate: 54% -> 60%Improved
add opponent modeling to AlphaStar (ablation)AlphaStar MA · Proxy VD opponentwin rate: 50% -> 56%Improved
add opponent modeling to AlphaStar (ablation)AlphaStar MA · Disruptor push opponentwin rate: 48% -> 58%Improved
scouting reward from opponent prediction (ablation)AlphaStar + opponent modeling · time to discover opponent buildingsdiscovery timeImproved
replace AlphaStar with full ROA-Star (goal-conditioned exploiters + opponent modeling)AlphaStar · 10-day leagueworst-case win rate: 1% -> 44%Improved
full ROA-Star vs AlphaStar average win rateAlphaStar · 10-day leagueaverage win rate: 70.2% -> 84.1%Improved
full ROA-Star vs AlphaStar RPPAlphaStar · 10-day leagueRPP score: 0.2157 -> 0.7843Improved
train exploiters on random z instead of unconditioned (ablation)AlphaStar · 5-day MA robustnessworst-case win rate / EloImproved
use goal-conditioned EIE+ERE instead of random-z exploiters (ablation)AlphaStar + random-z exploiters · 5-day MA robustnessworst-case win rate / EloImproved
add EIE on top of ERE-only exploiters (ablation)AlphaStar + ERE · 5-day MA EloEloImproved
use curricular reward and strategy-guided loss in ERE instead of edit-distance reward (ablation)Explorative Exploiters · target build order executionlast-entity execution precisionImproved

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