SA-IGA: A Multiagent Reinforcement Learning Method Towards Socially Optimal Outcomes

نویسندگان

  • Chengwei Zhang
  • Xiaohong Li
  • Jianye Hao
  • Siqi Chen
  • Karl Tuyls
  • Wanli Xue
چکیده

In multiagent environments, the capability of learning is important for an agent to behave appropriately in face of unknown opponents and dynamic environment. From the system designer’s perspective, it is desirable if the agents can learn to coordinate towards socially optimal outcomes, while also avoiding being exploited by selfish opponents. To this end, we propose a novel gradient ascent based algorithm (SA-IGA) which augments the basic gradient-ascent algorithm by incorporating social awareness into the policy update process. We theoretically analyze the learning dynamics of SA-IGA using dynamical system theory and SA-IGA is shown to have linear dynamics for a wide range of games including symmetric games. The learning dynamics of two representative games (the prisoner’s dilemma game and the coordination game) are analyzed in details. Based on the idea of SA-IGA, we further propose a practical multiagent learning algorithm, called SA-PGA, based on Q-learning update rule. Simulation results show that SA-PGA agent can achieve higher social welfare than previous social-optimality oriented Conditional Joint Action Learner (CJAL) and also is robust against individually rational opponents by reaching Nash equilibrium solutions.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Socially-Aware Multiagent Learning: Towards Socially Optimal Outcomes

In multiagent systems the capability of learning is important for an agent to behave appropriately in face of unknown opponents and a dynamic environment. From the system designer’s perspective, it is desirable if the agents can learn to coordinate towards socially optimal outcomes, while also avoiding being exploited by selfish opponents. To this end, we propose a novel gradient ascent based a...

متن کامل

A Multiagent Reinforcement Learning algorithm to solve the Community Detection Problem

Community detection is a challenging optimization problem that consists of searching for communities that belong to a network under the assumption that the nodes of the same community share properties that enable the detection of new characteristics or functional relationships in the network. Although there are many algorithms developed for community detection, most of them are unsuitable when ...

متن کامل

A Survey on Multiagent Reinforcement Learning Towards Multi-Robot Systems

Multiagent reinforcement learning for multirobot systems is a challenging issue in both robotics and artificial intelligence. With the ever increasing interests in theoretical research and practical applications, currently there have been a lot of efforts towards providing some solutions to this challenge. However, there are still many difficulties in scaling up multiagent reinforcement learnin...

متن کامل

Multiagent Reinforcement Learning: Theoretical Framework and an Algorithm

In this paper we adopt general sum stochas tic games as a framework for multiagent re inforcement learning Our work extends pre vious work by Littman on zero sum stochas tic games to a broader framework We de sign a multiagent Q learning method under this framework and prove that it converges to a Nash equilibrium under speci ed condi tions This algorithm is useful for nding the optimal strateg...

متن کامل

A multiagent architecture for concurrent reinforcement learning

In this paper we propose a multiagent architecture for implementing concurrent reinforcement learning, an approach where several agents, sharing the same environment, perceptions and actions, work towards one only objective: learning a single value function. We present encouraging experimental results derived from the initial phase of our research on the combination of concurrent reinforcement ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2018