Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/IQL

IQL

Reported on 42 benchmarks across 2 tasks · 2 papers

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology25 results

Multi-agent Reinforcement LearningonDef_Infantry_parallel
Median Win Rate· 2022-07-05
40
best: 100 (QTRAN)
The StarCraft Multi-Agent Challenges+ : Learning of Multi-Stage Tasks and Environmental Factors without Precise Reward Functions arXiv:2207.02007
Multi-agent Reinforcement LearningonDef_Armored_sequential
Median Win Rate· 2022-07-05
9.4
best: 100 (DRIMA)
The StarCraft Multi-Agent Challenges+ : Learning of Multi-Stage Tasks and Environmental Factors without Precise Reward Functions arXiv:2207.02007
Multi-agent Reinforcement LearningonDef_Infantry_sequential
Median Win Rate· 2022-07-05
93.8
best: 100 (MADDPG)
The StarCraft Multi-Agent Challenges+ : Learning of Multi-Stage Tasks and Environmental Factors without Precise Reward Functions arXiv:2207.02007
Multi-agent Reinforcement LearningonSMAC 3s5z_vs_3s6z
Average Score· 2021-02-16
16.54
best: 20.94 (DDN)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
Multi-agent Reinforcement LearningonSMAC 3s5z_vs_3s6z
Median Win Rate· 2021-02-16
29.83
best: 100 (ACE)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
Multi-agent Reinforcement LearningonSMAC corridor
Average Score· 2021-02-16
19.42
best: 20 (DDN)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
Multi-agent Reinforcement LearningonSMAC corridor
Median Win Rate· 2021-02-16
84.87
best: 100 (ACE)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
Multi-agent Reinforcement LearningonSMAC MMM2
Average Score· 2021-02-16
17.5
best: 20.9 (DDN)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
Multi-agent Reinforcement LearningonSMAC MMM2
Median Win Rate· 2021-02-16
68.92
best: 100 (ACE)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
Multi-agent Reinforcement LearningonSMAC 6h_vs_8z
Average Score· 2021-02-16
13.78
best: 19.4 (DDN)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
Multi-agent Reinforcement LearningonSMAC 27m_vs_30m
Average Score· 2021-02-16
14.01
best: 19.71 (DDN)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
Multi-agent Reinforcement LearningonSMAC 27m_vs_30m
Median Win Rate· 2021-02-16
2.27
best: 91.48 (DDN)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
Multi-agent Reinforcement LearningonSMAC 3s5z_vs_3s6z
Median Win Rate
0
best: 100 (ACE)
Multi-agent Reinforcement LearningonSMAC corridor
Median Win Rate
0
best: 100 (ACE)
Multi-agent Reinforcement LearningonSMAC MMM2
Median Win Rate
0
best: 100 (ACE)
Multi-agent Reinforcement LearningonSMAC 6h_vs_8z
Median Win Rate
0
best: 93.75 (ACE)
Multi-agent Reinforcement LearningonSMAC 6h_vs_8z
Median Win Rate
0
best: 93.75 (ACE)
Multi-agent Reinforcement LearningonOff_Hard_parallel
Median Win Rate
0
best: 80 (DRIMA)
Multi-agent Reinforcement LearningonDef_Outnumbered_sequential
Median Win Rate
0
best: 100 (DRIMA)
Multi-agent Reinforcement LearningonOff_Complicated_parallel
Median Win Rate
35
best: 100 (DRIMA)
Multi-agent Reinforcement LearningonOff_Near_parallel
Median Win Rate
5
best: 95 (QMIX)
Multi-agent Reinforcement LearningonDef_Armored_parallel
Median Win Rate
0
best: 90 (DMIX)
Multi-agent Reinforcement LearningonOff_Distant_parallel
Median Win Rate
0
best: 95 (DRIMA)
Multi-agent Reinforcement LearningonDef_Outnumbered_parallel
Median Win Rate
0
best: 70 (DRIMA)
Multi-agent Reinforcement LearningonOff_Superhard_parallel
Median Win Rate
0

Playing Games25 results

SMAConDef_Infantry_parallel
Median Win Rate· 2022-07-05
40
best: 100 (QTRAN)
The StarCraft Multi-Agent Challenges+ : Learning of Multi-Stage Tasks and Environmental Factors without Precise Reward Functions arXiv:2207.02007
SMAConDef_Armored_sequential
Median Win Rate· 2022-07-05
9.4
best: 100 (DRIMA)
The StarCraft Multi-Agent Challenges+ : Learning of Multi-Stage Tasks and Environmental Factors without Precise Reward Functions arXiv:2207.02007
SMAConDef_Infantry_sequential
Median Win Rate· 2022-07-05
93.8
best: 100 (MADDPG)
The StarCraft Multi-Agent Challenges+ : Learning of Multi-Stage Tasks and Environmental Factors without Precise Reward Functions arXiv:2207.02007
SMAConSMAC 3s5z_vs_3s6z
Average Score· 2021-02-16
16.54
best: 20.94 (DDN)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
SMAConSMAC 3s5z_vs_3s6z
Median Win Rate· 2021-02-16
29.83
best: 100 (ACE)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
SMAConSMAC corridor
Average Score· 2021-02-16
19.42
best: 20 (DDN)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
SMAConSMAC corridor
Median Win Rate· 2021-02-16
84.87
best: 100 (ACE)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
SMAConSMAC MMM2
Average Score· 2021-02-16
17.5
best: 20.9 (DDN)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
SMAConSMAC MMM2
Median Win Rate· 2021-02-16
68.92
best: 100 (ACE)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
SMAConSMAC 6h_vs_8z
Average Score· 2021-02-16
13.78
best: 19.4 (DDN)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
SMAConSMAC 27m_vs_30m
Average Score· 2021-02-16
14.01
best: 19.71 (DDN)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
SMAConSMAC 27m_vs_30m
Median Win Rate· 2021-02-16
2.27
best: 91.48 (DDN)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning arXiv:2102.07936
SMAConSMAC 3s5z_vs_3s6z
Median Win Rate
0
best: 100 (ACE)
SMAConSMAC corridor
Median Win Rate
0
best: 100 (ACE)
SMAConSMAC MMM2
Median Win Rate
0
best: 100 (ACE)
SMAConSMAC 6h_vs_8z
Median Win Rate
0
best: 93.75 (ACE)
SMAConSMAC 6h_vs_8z
Median Win Rate
0
best: 93.75 (ACE)
SMAConOff_Hard_parallel
Median Win Rate
0
best: 80 (DRIMA)
SMAConDef_Outnumbered_sequential
Median Win Rate
0
best: 100 (DRIMA)
SMAConOff_Complicated_parallel
Median Win Rate
35
best: 100 (DRIMA)
SMAConOff_Near_parallel
Median Win Rate
5
best: 95 (QMIX)
SMAConDef_Armored_parallel
Median Win Rate
0
best: 90 (DMIX)
SMAConOff_Distant_parallel
Median Win Rate
0
best: 95 (DRIMA)
SMAConDef_Outnumbered_parallel
Median Win Rate
0
best: 70 (DRIMA)
SMAConOff_Superhard_parallel
Median Win Rate
0