Generating interpretable fuzzy controllers using particle swarm optimization and genetic programming

Created by W.Langdon from gp-bibliography.bib Revision:1.7630

@InProceedings{Hein:2018:GECCOcomp,

• author = "Daniel Hein and Steffen Udluft and Thomas A. Runkler",
• title = "Generating interpretable fuzzy controllers using particle swarm optimization and genetic programming",
• booktitle = "GECCO '18: Proceedings of the Genetic and Evolutionary Computation Conference Companion",
• year = "2018",
• editor = "Carlos Cotta and Tapabrata Ray and Hisao Ishibuchi and Shigeru Obayashi and Bogdan Filipic and Thomas Bartz-Beielstein and Grant Dick and Masaharu Munetomo and Silvino {Fernandez Alzueta} and Thomas Stuetzle and Pablo Valledor Pellicer and Manuel Lopez-Ibanez and Daniel R. Tauritz and Pietro S. Oliveto and Thomas Weise and Borys Wrobel and Ales Zamuda and Anne Auger and Julien Bect and Dimo Brockhoff and Nikolaus Hansen and Rodolphe {Le Riche} and Victor Picheny and Bilel Derbel and Ke Li and Hui Li and Xiaodong Li and Saul Zapotecas and Qingfu Zhang and Stephane Doncieux and Richard Duro and Joshua Auerbach and Harold {de Vladar} and Antonio J. Fernandez-Leiva and JJ Merelo and Pedro A. Castillo-Valdivieso and David Camacho-Fernandez and Francisco {Chavez de la O} and Ozgur Akman and Khulood Alyahya and Juergen Branke and Kevin Doherty and Jonathan Fieldsend and Giuseppe Carlo Marano and Nikos D. Lagaros and Koichi Nakayama and Chika Oshima and Stefan Wagner and Michael Affenzeller and Boris Naujoks and Vanessa Volz and Tea Tusar and Pascal Kerschke and Riyad Alshammari and Tokunbo Makanju and Brad Alexander and Saemundur O. Haraldsson and Markus Wagner and John R. Woodward and Shin Yoo and John McCall and Nayat Sanchez-Pi and Luis Marti and Danilo Vasconcellos and Masaya Nakata and Anthony Stein and Nadarajen Veerapen and Arnaud Liefooghe and Sebastien Verel and Gabriela Ochoa and Stephen L. Smith and Stefano Cagnoni and Robert M. Patton and William {La Cava} and Randal Olson and Patryk Orzechowski and Ryan Urbanowicz and Ivanoe {De Falco} and Antonio {Della Cioppa} and Ernesto Tarantino and Umberto Scafuri and P. G. M. Baltus and Giovanni Iacca and Ahmed Hallawa and Anil Yaman and Alma Rahat and Handing Wang and Yaochu Jin and David Walker and Richard Everson and Akira Oyama and Koji Shimoyama and Hemant Kumar and Kazuhisa Chiba and Pramudita Satria Palar",
• isbn13 = "978-1-4503-5764-7",
• pages = "1268--1275",
• address = "Kyoto, Japan",
• publisher = "ACM",
• publisher_address = "New York, NY, USA",
• month = "15-19 " # jul,
• organisation = "SIGEVO",
• keywords = "genetic algorithms, genetic programming, interpretable reinforcement learning, RL, fuzzy control, swarm optimization, PSO, FGPRL, FGPRL, AMIFS, industrial benchmark",
• URL = "https://arxiv.org/abs/1804.10960",
• DOI = "doi:10.1145/3205651.3208277",
• size = "8 pages",
• abstract = "Autonomously training interpretable control strategies, called policies, using pre-existing plant trajectory data is of great interest in industrial applications. Fuzzy controllers have been used in industry for decades as interpretable and efficient system controllers. In this study, we introduce a fuzzy genetic programming (GP) approach called fuzzy GP reinforcement learning (FGPRL) that can select the relevant state features, determine the size of the required fuzzy rule set, and automatically adjust all the controller parameters simultaneously. Each GP individual's fitness is computed using model-based batch reinforcement learning (RL), which first trains a model using available system samples and subsequently performs Monte Carlo roll outs to predict each policy candidate's performance. We compare FGPRL to an extended version of a related method called fuzzy particle swarm reinforcement learning (FPSRL), which uses swarm intelligence to tune the fuzzy policy parameters. Experiments using an industrial benchmark show that FGPRL is able to autonomously learn interpretable fuzzy policies with high control performance.",
• notes = "Fig. 2. A fuzzy GP individual

  Table 1: Types and complexity. Local Search.

  'GP-based fuzzy policy learning approach called FGPRL ... FGPRL is better than FPSRL at creating interpretable fuzzy policies autonomously from existing transition samples'

  http://github.com/siemens/industrialbenchmark

  Siemens AG, Corporate Technology, Germany

  Also known as \cite{3208277} GECCO-2018 A Recombination of the 27th International Conference on Genetic Algorithms (ICGA-2018) and the 23rd Annual Genetic Programming Conference (GP-2018)",

}

Genetic Programming entries for Daniel Hein Steffen Udluft Thomas A Runkler

Citations