Reinforcement-driven adaptation of control relations

Hans Arno Jacobsen; Joachim Weisbrod

Reinforcement-driven adaptation of control relations

Hans Arno Jacobsen, Joachim Weisbrod

University of California at Berkeley

Research output: Contribution to conference › Paper › peer-review

1 Scopus citations

Abstract

The conceptual framework of a hybrid control system architecture is briefly motivated. It employs neural and fuzzy techniques on a side-by-side basis using each one for the task it is best suited for. In this paper, our main interest is with the adaptation of the fuzzy control knowledge. The adaptation algorithm is based on reinforcement signals and directly optimizes the global fuzzy relation representing the complete knowledge base. The new approach is experimentally evaluated.

Original language	English
Pages	484-488
Number of pages	5
State	Published - 1996
Externally published	Yes
Event	Proceedings of the 1996 Biennial Conference of the North American Fuzzy Information Processing Society - NAFIPS - Berkeley, CA, USA Duration: 19 Jun 1996 → 22 Jun 1996

Conference

Conference	Proceedings of the 1996 Biennial Conference of the North American Fuzzy Information Processing Society - NAFIPS
City	Berkeley, CA, USA
Period	19/06/96 → 22/06/96

Cite this

@conference{f2066350b775451fbbe4d56a7b4262fa,

title = "Reinforcement-driven adaptation of control relations",

abstract = "The conceptual framework of a hybrid control system architecture is briefly motivated. It employs neural and fuzzy techniques on a side-by-side basis using each one for the task it is best suited for. In this paper, our main interest is with the adaptation of the fuzzy control knowledge. The adaptation algorithm is based on reinforcement signals and directly optimizes the global fuzzy relation representing the complete knowledge base. The new approach is experimentally evaluated.",

author = "Jacobsen, \{Hans Arno\} and Joachim Weisbrod",

year = "1996",

language = "English",

pages = "484--488",

note = "Proceedings of the 1996 Biennial Conference of the North American Fuzzy Information Processing Society - NAFIPS ; Conference date: 19-06-1996 Through 22-06-1996",

}

TY - CONF

T1 - Reinforcement-driven adaptation of control relations

AU - Jacobsen, Hans Arno

AU - Weisbrod, Joachim

PY - 1996

Y1 - 1996

N2 - The conceptual framework of a hybrid control system architecture is briefly motivated. It employs neural and fuzzy techniques on a side-by-side basis using each one for the task it is best suited for. In this paper, our main interest is with the adaptation of the fuzzy control knowledge. The adaptation algorithm is based on reinforcement signals and directly optimizes the global fuzzy relation representing the complete knowledge base. The new approach is experimentally evaluated.

AB - The conceptual framework of a hybrid control system architecture is briefly motivated. It employs neural and fuzzy techniques on a side-by-side basis using each one for the task it is best suited for. In this paper, our main interest is with the adaptation of the fuzzy control knowledge. The adaptation algorithm is based on reinforcement signals and directly optimizes the global fuzzy relation representing the complete knowledge base. The new approach is experimentally evaluated.

UR - http://www.scopus.com/inward/record.url?scp=0029697396&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0029697396

SP - 484

EP - 488

T2 - Proceedings of the 1996 Biennial Conference of the North American Fuzzy Information Processing Society - NAFIPS

Y2 - 19 June 1996 through 22 June 1996

ER -

Reinforcement-driven adaptation of control relations

Abstract

Conference

Other files and links

Fingerprint

Cite this