DISTRIBUTED REINFORCEMENT LEARNING FOR A TRAFFIC ENGINEERING APPLICATION

In this paper, the authors describe how a distributed reinforcement learning problem, in which the returns of many agents are simultaneously updating a single shared policy, is addressed by applying novel reinforcement learning techniques. A traffic simulator is used in the learning process. Two new algorithms are introduced: a value function-based algorithm and one that uses a direct policy evaluation approach. Both algorithms are shown to perform comparably well.

• Supplemental Notes:
  • Publication Date: 2000. Association for Computing Machinery, New York NY
• Corporate Authors:

  DaimlerChrysler Research and Technology Center North America

  ,    
• Authors:
  • Pendrith, M D
• Conference:
  • International Conference on Autonomous Agents (4th : 2000 : Barcelona Spain). Proceedings
• Publication Date: 2000

Language

• English

Media Info

• Pagination: p. 404-411

Subject/Index Terms

• TRT Terms: Artificial intelligence; Traffic control
• Subject Areas: Operations and Traffic Management;

Filing Info

• Accession Number: 00821745
• Record Type: Publication
• Source Agency: UC Berkeley Transportation Library
• Files: PATH
• Created Date: Dec 31 2001 12:00AM