Fuzzy policy gradient reinforcement learning for leader-follower systems

Dongbing Gu, Erfu Yang

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

2 Citations (Scopus)

Abstract

This paper presents a policy gradient multi-agent reinforcement learning algorithm for leader-follower systems. In this algorithm, cooperative dynamics of the leader-follower control is modelled as an incentive Stackelberg game. A linear incentive mechanism is used to connect the leader and follower policies. Policy gradient reinforcement learning explicitly explores policy parameter space to search the optimal policy. Fuzzy logic controllers are used as the policy. The parameters of fuzzy logic controllers can be improved by this policy gradient algorithm.

Original languageEnglish
Title of host publication2005 IEEE International Conference on Mechatronics & Automations
Subtitle of host publicationConference Proceedings
EditorsJason Gu, Peter X. Liu
Place of PublicationPiscataway, NJ.
PublisherIEEE
Pages1557-1561
Number of pages5
Volume3
ISBN (Print)078039044X
DOIs
Publication statusPublished - 1 Jul 2005
EventIEEE International Conference on Mechatronics and Automation, ICMA 2005 - Niagara Falls, ON, United Kingdom
Duration: 29 Jul 20051 Aug 2005

Conference

ConferenceIEEE International Conference on Mechatronics and Automation, ICMA 2005
CountryUnited Kingdom
CityNiagara Falls, ON
Period29/07/051/08/05

Fingerprint

Reinforcement learning
Fuzzy logic
Controllers
Learning algorithms

Keywords

  • incentive Stackelberg game
  • multi-agent reinforcement learning
  • policy gradient reinforcement learning
  • control engineering computing
  • fuzzy logic
  • game theory
  • learning (artificial intelligence)
  • multi-agent systems

Cite this

Gu, D., & Yang, E. (2005). Fuzzy policy gradient reinforcement learning for leader-follower systems. In J. Gu, & P. X. Liu (Eds.), 2005 IEEE International Conference on Mechatronics & Automations : Conference Proceedings (Vol. 3, pp. 1557-1561). Piscataway, NJ.: IEEE. https://doi.org/10.1109/ICMA.2005.1626787
Gu, Dongbing ; Yang, Erfu. / Fuzzy policy gradient reinforcement learning for leader-follower systems. 2005 IEEE International Conference on Mechatronics & Automations : Conference Proceedings. editor / Jason Gu ; Peter X. Liu. Vol. 3 Piscataway, NJ. : IEEE, 2005. pp. 1557-1561
@inproceedings{dd4f0eb28f4c4878a66f340ccdfff9bf,
title = "Fuzzy policy gradient reinforcement learning for leader-follower systems",
abstract = "This paper presents a policy gradient multi-agent reinforcement learning algorithm for leader-follower systems. In this algorithm, cooperative dynamics of the leader-follower control is modelled as an incentive Stackelberg game. A linear incentive mechanism is used to connect the leader and follower policies. Policy gradient reinforcement learning explicitly explores policy parameter space to search the optimal policy. Fuzzy logic controllers are used as the policy. The parameters of fuzzy logic controllers can be improved by this policy gradient algorithm.",
keywords = "incentive Stackelberg game, multi-agent reinforcement learning, policy gradient reinforcement learning, control engineering computing, fuzzy logic, game theory, learning (artificial intelligence), multi-agent systems",
author = "Dongbing Gu and Erfu Yang",
year = "2005",
month = "7",
day = "1",
doi = "10.1109/ICMA.2005.1626787",
language = "English",
isbn = "078039044X",
volume = "3",
pages = "1557--1561",
editor = "Jason Gu and Liu, {Peter X.}",
booktitle = "2005 IEEE International Conference on Mechatronics & Automations",
publisher = "IEEE",

}

Gu, D & Yang, E 2005, Fuzzy policy gradient reinforcement learning for leader-follower systems. in J Gu & PX Liu (eds), 2005 IEEE International Conference on Mechatronics & Automations : Conference Proceedings. vol. 3, IEEE, Piscataway, NJ., pp. 1557-1561, IEEE International Conference on Mechatronics and Automation, ICMA 2005, Niagara Falls, ON, United Kingdom, 29/07/05. https://doi.org/10.1109/ICMA.2005.1626787

Fuzzy policy gradient reinforcement learning for leader-follower systems. / Gu, Dongbing; Yang, Erfu.

2005 IEEE International Conference on Mechatronics & Automations : Conference Proceedings. ed. / Jason Gu; Peter X. Liu. Vol. 3 Piscataway, NJ. : IEEE, 2005. p. 1557-1561.

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

TY - GEN

T1 - Fuzzy policy gradient reinforcement learning for leader-follower systems

AU - Gu, Dongbing

AU - Yang, Erfu

PY - 2005/7/1

Y1 - 2005/7/1

N2 - This paper presents a policy gradient multi-agent reinforcement learning algorithm for leader-follower systems. In this algorithm, cooperative dynamics of the leader-follower control is modelled as an incentive Stackelberg game. A linear incentive mechanism is used to connect the leader and follower policies. Policy gradient reinforcement learning explicitly explores policy parameter space to search the optimal policy. Fuzzy logic controllers are used as the policy. The parameters of fuzzy logic controllers can be improved by this policy gradient algorithm.

AB - This paper presents a policy gradient multi-agent reinforcement learning algorithm for leader-follower systems. In this algorithm, cooperative dynamics of the leader-follower control is modelled as an incentive Stackelberg game. A linear incentive mechanism is used to connect the leader and follower policies. Policy gradient reinforcement learning explicitly explores policy parameter space to search the optimal policy. Fuzzy logic controllers are used as the policy. The parameters of fuzzy logic controllers can be improved by this policy gradient algorithm.

KW - incentive Stackelberg game

KW - multi-agent reinforcement learning

KW - policy gradient reinforcement learning

KW - control engineering computing

KW - fuzzy logic

KW - game theory

KW - learning (artificial intelligence)

KW - multi-agent systems

UR - http://www.scopus.com/inward/record.url?scp=27744514981&partnerID=8YFLogxK

UR - http://ieeexplore.ieee.org/xpl/tocresult.jsp?isnumber=34148

UR - http://myweb.dal.ca/jgu/icma05/

U2 - 10.1109/ICMA.2005.1626787

DO - 10.1109/ICMA.2005.1626787

M3 - Conference contribution book

SN - 078039044X

VL - 3

SP - 1557

EP - 1561

BT - 2005 IEEE International Conference on Mechatronics & Automations

A2 - Gu, Jason

A2 - Liu, Peter X.

PB - IEEE

CY - Piscataway, NJ.

ER -

Gu D, Yang E. Fuzzy policy gradient reinforcement learning for leader-follower systems. In Gu J, Liu PX, editors, 2005 IEEE International Conference on Mechatronics & Automations : Conference Proceedings. Vol. 3. Piscataway, NJ.: IEEE. 2005. p. 1557-1561 https://doi.org/10.1109/ICMA.2005.1626787