HyperAI초신경

홈 뉴스 최신 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Smac On Smac Corridor

평가 지표

Average Score

Median Win Rate

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	Average Score	Median Win Rate	Paper Title	Repository
DIQL	19.68	91.62	DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning
DPLEX	19.08	81.25	A Unified Framework for Factorizing Distributional Value Functions for Multi-Agent Reinforcement Learning
VDN	19.47	85.34	DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning
IQL	-	0	The StarCraft Multi-Agent Challenge
DDN	20	95.4	DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning
Heuristic	-	0	The StarCraft Multi-Agent Challenge
QMIX	15.07	37.61	DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning
DMIX	19.66	90.45	DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning
QPLEX	18.73	75.00	A Unified Framework for Factorizing Distributional Value Functions for Multi-Agent Reinforcement Learning
ACE	-	100	ACE: Cooperative Multi-agent Q-learning with Bidirectional Action-Dependency
QMIX	-	1	Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning
QMIX	-	1	The StarCraft Multi-Agent Challenge
IQL	19.42	84.87	DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning

0 of 13 row(s) selected.