Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Yaodong Yang

Yaodong Yang

Online Markov Decision Processes with Non-oblivious Strategic Adversary


Oct 08, 2021
Le Cong Dinh, David Henry Mguni, Long Tran-Thanh, Jun Wang, Yaodong Yang

Figure 1 for Online Markov Decision Processes with Non-oblivious Strategic Adversary

  Access Paper or Ask Questions

Multi-Agent Constrained Policy Optimisation


Oct 06, 2021
Shangding Gu, Jakub Grudzien Kuba, Munning Wen, Ruiqing Chen, Ziyan Wang, Zheng Tian, Jun Wang, Alois Knoll, Yaodong Yang

Figure 1 for Multi-Agent Constrained Policy Optimisation
Figure 2 for Multi-Agent Constrained Policy Optimisation
Figure 3 for Multi-Agent Constrained Policy Optimisation
Figure 4 for Multi-Agent Constrained Policy Optimisation

  Access Paper or Ask Questions

Trust Region Policy Optimisation in Multi-Agent Reinforcement Learning


Sep 23, 2021
Jakub Grudzien Kuba, Ruiqing Chen, Munning Wen, Ying Wen, Fanglei Sun, Jun Wang, Yaodong Yang

Figure 1 for Trust Region Policy Optimisation in Multi-Agent Reinforcement Learning
Figure 2 for Trust Region Policy Optimisation in Multi-Agent Reinforcement Learning
Figure 3 for Trust Region Policy Optimisation in Multi-Agent Reinforcement Learning
Figure 4 for Trust Region Policy Optimisation in Multi-Agent Reinforcement Learning

  Access Paper or Ask Questions

Revisiting the Characteristics of Stochastic Gradient Noise and Dynamics


Sep 20, 2021
Yixin Wu, Rui Luo, Chen Zhang, Jun Wang, Yaodong Yang

Figure 1 for Revisiting the Characteristics of Stochastic Gradient Noise and Dynamics
Figure 2 for Revisiting the Characteristics of Stochastic Gradient Noise and Dynamics
Figure 3 for Revisiting the Characteristics of Stochastic Gradient Noise and Dynamics
Figure 4 for Revisiting the Characteristics of Stochastic Gradient Noise and Dynamics
* 18 pages 
  Access Paper or Ask Questions

On the Complexity of Computing Markov Perfect Equilibrium in General-Sum Stochastic Games


Sep 04, 2021
Xiaotie Deng, Yuhao Li, David Henry Mguni, Jun Wang, Yaodong Yang


  Access Paper or Ask Questions

Settling the Variance of Multi-Agent Policy Gradients


Aug 20, 2021
Jakub Grudzien Kuba, Muning Wen, Yaodong Yang, Linghui Meng, Shangding Gu, Haifeng Zhang, David Henry Mguni, Jun Wang

Figure 1 for Settling the Variance of Multi-Agent Policy Gradients
Figure 2 for Settling the Variance of Multi-Agent Policy Gradients
Figure 3 for Settling the Variance of Multi-Agent Policy Gradients

  Access Paper or Ask Questions

A Game-Theoretic Approach to Multi-Agent Trust Region Optimization


Jun 12, 2021
Ying Wen, Hui Chen, Yaodong Yang, Zheng Tian, Minne Li, Xu Chen, Jun Wang

Figure 1 for A Game-Theoretic Approach to Multi-Agent Trust Region Optimization
Figure 2 for A Game-Theoretic Approach to Multi-Agent Trust Region Optimization
Figure 3 for A Game-Theoretic Approach to Multi-Agent Trust Region Optimization
Figure 4 for A Game-Theoretic Approach to Multi-Agent Trust Region Optimization
* A Multi-Agent Trust Region Learning (MATRL) algorithm that augments the single-agent trust region policy optimization with a weak stable fixed point approximated by the policy-space meta-game 
  Access Paper or Ask Questions

Unifying Behavioral and Response Diversity for Open-ended Learning in Zero-sum Games


Jun 10, 2021
Xiangyu Liu, Hangtian Jia, Ying Wen, Yaodong Yang, Yujing Hu, Yingfeng Chen, Changjie Fan, Zhipeng Hu

Figure 1 for Unifying Behavioral and Response Diversity for Open-ended Learning in Zero-sum Games
Figure 2 for Unifying Behavioral and Response Diversity for Open-ended Learning in Zero-sum Games
Figure 3 for Unifying Behavioral and Response Diversity for Open-ended Learning in Zero-sum Games
Figure 4 for Unifying Behavioral and Response Diversity for Open-ended Learning in Zero-sum Games
* We investigate a new perspective on unifying diversity measures for open-ended learning in zero-sum games, which shapes an auto-curriculum to induce diverse yet effective behaviors 
  Access Paper or Ask Questions

MALib: A Parallel Framework for Population-based Multi-agent Reinforcement Learning


Jun 05, 2021
Ming Zhou, Ziyu Wan, Hanjing Wang, Muning Wen, Runzhe Wu, Ying Wen, Yaodong Yang, Weinan Zhang, Jun Wang

Figure 1 for MALib: A Parallel Framework for Population-based Multi-agent Reinforcement Learning
Figure 2 for MALib: A Parallel Framework for Population-based Multi-agent Reinforcement Learning
Figure 3 for MALib: A Parallel Framework for Population-based Multi-agent Reinforcement Learning
Figure 4 for MALib: A Parallel Framework for Population-based Multi-agent Reinforcement Learning
* 24 pages, 17 figures, 5 tables 
  Access Paper or Ask Questions

Discovering Multi-Agent Auto-Curricula in Two-Player Zero-Sum Games


Jun 04, 2021
Xidong Feng, Oliver Slumbers, Yaodong Yang, Ziyu Wan, Bo Liu, Stephen McAleer, Ying Wen, Jun Wang

Figure 1 for Discovering Multi-Agent Auto-Curricula in Two-Player Zero-Sum Games
Figure 2 for Discovering Multi-Agent Auto-Curricula in Two-Player Zero-Sum Games
Figure 3 for Discovering Multi-Agent Auto-Curricula in Two-Player Zero-Sum Games
Figure 4 for Discovering Multi-Agent Auto-Curricula in Two-Player Zero-Sum Games
* corresponding to  
  Access Paper or Ask Questions

Cooperative Multi-Agent Transfer Learning with Level-Adaptive Credit Assignment


Jun 03, 2021
Tianze Zhou, Fubiao Zhang, Kun Shao, Kai Li, Wenhan Huang, Jun Luo, Weixun Wang, Yaodong Yang, Hangyu Mao, Bin Wang, Dong Li, Wulong Liu, Jianye Hao

Figure 1 for Cooperative Multi-Agent Transfer Learning with Level-Adaptive Credit Assignment
Figure 2 for Cooperative Multi-Agent Transfer Learning with Level-Adaptive Credit Assignment
Figure 3 for Cooperative Multi-Agent Transfer Learning with Level-Adaptive Credit Assignment
Figure 4 for Cooperative Multi-Agent Transfer Learning with Level-Adaptive Credit Assignment
* 12 pages, 9 figures 
  Access Paper or Ask Questions

Learning to Shape Rewards using a Game of Switching Controls


Mar 16, 2021
David Mguni, Jianhong Wang, Taher Jafferjee, Nicolas Perez-Nieves, Wenbin Song, Yaodong Yang, Feifei Tong, Hui Chen, Jiangcheng Zhu, Yali Du, Jun Wang

Figure 1 for Learning to Shape Rewards using a Game of Switching Controls
Figure 2 for Learning to Shape Rewards using a Game of Switching Controls
Figure 3 for Learning to Shape Rewards using a Game of Switching Controls
Figure 4 for Learning to Shape Rewards using a Game of Switching Controls

  Access Paper or Ask Questions

Online Double Oracle


Mar 16, 2021
Le Cong Dinh, Yaodong Yang, Zheng Tian, Nicolas Perez Nieves, Oliver Slumbers, David Henry Mguni, Haitham Bou Ammar, Jun Wang

Figure 1 for Online Double Oracle
Figure 2 for Online Double Oracle
Figure 3 for Online Double Oracle
Figure 4 for Online Double Oracle
* [email protected] 
  Access Paper or Ask Questions

Modelling Behavioural Diversity for Learning in Open-Ended Games


Mar 14, 2021
Nicolas Perez Nieves, Yaodong Yang, Oliver Slumbers, David Henry Mguni, Jun Wang

Figure 1 for Modelling Behavioural Diversity for Learning in Open-Ended Games
Figure 2 for Modelling Behavioural Diversity for Learning in Open-Ended Games
Figure 3 for Modelling Behavioural Diversity for Learning in Open-Ended Games
Figure 4 for Modelling Behavioural Diversity for Learning in Open-Ended Games
* corresponds to  
  Access Paper or Ask Questions

Foresee then Evaluate: Decomposing Value Estimation with Latent Future Prediction


Mar 03, 2021
Hongyao Tang, Jianye Hao, Guangyong Chen, Pengfei Chen, Chen Chen, Yaodong Yang, Luo Zhang, Wulong Liu, Zhaopeng Meng

Figure 1 for Foresee then Evaluate: Decomposing Value Estimation with Latent Future Prediction
Figure 2 for Foresee then Evaluate: Decomposing Value Estimation with Latent Future Prediction
Figure 3 for Foresee then Evaluate: Decomposing Value Estimation with Latent Future Prediction
Figure 4 for Foresee then Evaluate: Decomposing Value Estimation with Latent Future Prediction
* Accepted paper on AAAI 2021. arXiv admin note: text overlap with arXiv:1905.11100 
  Access Paper or Ask Questions

Diverse Auto-Curriculum is Critical for Successful Real-World Multiagent Learning Systems


Feb 16, 2021
Yaodong Yang, Jun Luo, Ying Wen, Oliver Slumbers, Daniel Graves, Haitham Bou Ammar, Jun Wang, Matthew E. Taylor

* AAMAS 2021 
  Access Paper or Ask Questions

An Overview of Multi-Agent Reinforcement Learning from Game Theoretical Perspective


Nov 01, 2020
Yaodong Yang, Jun Wang

Figure 1 for An Overview of Multi-Agent Reinforcement Learning from Game Theoretical Perspective
Figure 2 for An Overview of Multi-Agent Reinforcement Learning from Game Theoretical Perspective
Figure 3 for An Overview of Multi-Agent Reinforcement Learning from Game Theoretical Perspective
Figure 4 for An Overview of Multi-Agent Reinforcement Learning from Game Theoretical Perspective

  Access Paper or Ask Questions

SMARTS: Scalable Multi-Agent Reinforcement Learning Training School for Autonomous Driving


Nov 01, 2020
Ming Zhou, Jun Luo, Julian Villella, Yaodong Yang, David Rusu, Jiayu Miao, Weinan Zhang, Montgomery Alban, Iman Fadakar, Zheng Chen, Aurora Chongxi Huang, Ying Wen, Kimia Hassanzadeh, Daniel Graves, Dong Chen, Zhengbang Zhu, Nhat Nguyen, Mohamed Elsayed, Kun Shao, Sanjeevan Ahilan, Baokuan Zhang, Jiannan Wu, Zhengang Fu, Kasra Rezaee, Peyman Yadmellat, Mohsen Rohani, Nicolas Perez Nieves, Yihan Ni, Seyedershad Banijamali, Alexander Cowen Rivers, Zheng Tian, Daniel Palenicek, Haitham bou Ammar, Hongbo Zhang, Wulong Liu, Jianye Hao, Jun Wang

Figure 1 for SMARTS: Scalable Multi-Agent Reinforcement Learning Training School for Autonomous Driving
Figure 2 for SMARTS: Scalable Multi-Agent Reinforcement Learning Training School for Autonomous Driving
Figure 3 for SMARTS: Scalable Multi-Agent Reinforcement Learning Training School for Autonomous Driving
Figure 4 for SMARTS: Scalable Multi-Agent Reinforcement Learning Training School for Autonomous Driving
* 20 pages, 11 figures. Paper accepted to CoRL 2020 
  Access Paper or Ask Questions

What About Taking Policy as Input of Value Function: Policy-extended Value Function Approximator


Oct 19, 2020
Hongyao Tang, Zhaopeng Meng, Jianye HAO, Chen Chen, Daniel Graves, Dong Li, Wulong Liu, Yaodong Yang

Figure 1 for What About Taking Policy as Input of Value Function: Policy-extended Value Function Approximator
Figure 2 for What About Taking Policy as Input of Value Function: Policy-extended Value Function Approximator
Figure 3 for What About Taking Policy as Input of Value Function: Policy-extended Value Function Approximator
Figure 4 for What About Taking Policy as Input of Value Function: Policy-extended Value Function Approximator
* Preprint version 
  Access Paper or Ask Questions

Learning to Infer User Hidden States for Online Sequential Advertising


Sep 03, 2020
Zhaoqing Peng, Junqi Jin, Lan Luo, Yaodong Yang, Rui Luo, Jun Wang, Weinan Zhang, Haiyang Xu, Miao Xu, Chuan Yu, Tiejian Luo, Han Li, Jian Xu, Kun Gai

Figure 1 for Learning to Infer User Hidden States for Online Sequential Advertising
Figure 2 for Learning to Infer User Hidden States for Online Sequential Advertising
Figure 3 for Learning to Infer User Hidden States for Online Sequential Advertising
Figure 4 for Learning to Infer User Hidden States for Online Sequential Advertising
* to be published in CIKM 2020 
  Access Paper or Ask Questions

Multi-Agent Determinantal Q-Learning


Jun 09, 2020
Yaodong Yang, Ying Wen, Liheng Chen, Jun Wang, Kun Shao, David Mguni, Weinan Zhang

Figure 1 for Multi-Agent Determinantal Q-Learning
Figure 2 for Multi-Agent Determinantal Q-Learning
Figure 3 for Multi-Agent Determinantal Q-Learning
Figure 4 for Multi-Agent Determinantal Q-Learning
* ICML 2020 
  Access Paper or Ask Questions

$α^α$-Rank: Scalable Multi-agent Evaluation through Evolution


Sep 28, 2019
Yaodong Yang, Rasul Tutunov, Phu Sakulwongtana, Haitham Bou Ammar, Jun Wang

Figure 1 for $α^α$-Rank: Scalable Multi-agent Evaluation through Evolution
Figure 2 for $α^α$-Rank: Scalable Multi-agent Evaluation through Evolution
Figure 3 for $α^α$-Rank: Scalable Multi-agent Evaluation through Evolution
Figure 4 for $α^α$-Rank: Scalable Multi-agent Evaluation through Evolution
* The authors decide to retract the current version for an improvement 
  Access Paper or Ask Questions