Risk-sensitive optimal control for stochastic recurrent neural networks

Author

Liu, Ziqian ; Torres, Raul E. ; Kotinis, Miltiadis

Author_Institution

Eng. Dept., State Univ. of New York Maritime Coll., Throggs Neck, NY, USA

fYear

2010

Firstpage

1137

Lastpage

1140

Abstract

As a continuation of our study, this paper extends our research results of optimality-oriented control from deterministic recurrent neural networks to stochastic recurrent neural networks, and presents a new theoretical design for the risk-sensitive optimal control of stochastic recurrent neural networks. The design procedure follows the technique of inverse optimality, and obtains risk-sensitive state feedback controllers that guarantee an achievable meaningful cost for a given risk-sensitivity parameter.

Keywords

nonlinear control systems; optimal control; recurrent neural nets; state feedback; stochastic systems; deterministic recurrent neural networks; inverse optimality technique; risk-sensitive optimal control; risk-sensitive state feedback controllers; risk-sensitivity parameter; stochastic recurrent neural networks; Control systems; Controllability; Cost function; Nonlinear control systems; Nonlinear systems; Optimal control; Recurrent neural networks; Stochastic processes; Stochastic resonance; System identification;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems (MWSCAS), 2010 53rd IEEE International Midwest Symposium on

Conference_Location

Seattle, WA

ISSN

1548-3746

Print_ISBN

978-1-4244-7771-5

Type

conf

DOI

10.1109/MWSCAS.2010.5548858

Filename

5548858