SPSA with a fixed gain for intelligent control in tracking applications

Simultaneous perturbation stochastic approximation (SPSA) algorithm is also often referred as a Kiefer-Wolfowitz algorithm with randomized differences. Algorithms of this type are widely applied in field of intelligent control for optimization purposes, especially in a high-dimensional and noisy setting. In such problems it is often important to track the drifting minimum point, adapting to changing environment. In this paper application of the fixed gain SPSA to the minimum tracking problem for the non-constrained optimization is considered. The upper bound of mean square estimation error is determined in case of once differentiable functional and almost arbitrary noises. Numerical simulation of the estimates stabilization for the multidimensional optimization with non-random noise is provided.

[1]  Han-Fu Chen,et al.  A Kiefer-Wolfowitz algorithm with randomized differences , 1999, IEEE Trans. Autom. Control..

[2]  A. Yu. Popkov,et al.  Gradient Methods for Nonstationary Unconstrained Optimization Problems , 2005 .

[3]  Jiashun Zhang,et al.  Fuzzy age-dependent replacement policy and SPSA algorithm based-on fuzzy simulation , 2008, Inf. Sci..

[4]  H. Kushner,et al.  Stochastic Approximation and Recursive Algorithms and Applications , 2003 .

[5]  Rui J. P. de Figueiredo,et al.  Learning rules for neuro-controller via simultaneous perturbation , 1997, IEEE Trans. Neural Networks.

[6]  László Gerencsér,et al.  Optimization over discrete sets via SPSA , 1999, WSC '99.

[7]  V. Borkar.  Stochastic Approximation: A Dynamical Systems Viewpoint , 2008 .

[8]  Lei Guo,et al.  Stability of Recursive Stochastic Tracking Algorithms , 1994 .

[9]  Oleg Granichin,et al.  Linear regression and filtering under nonstandard assumptions (arbitrary noise) , 2004, IEEE Transactions on Automatic Control.

[10]  Oleg Granichin,et al.  A stochastic recursive procedure with dependent noises in the observation that uses sample perturbations in the input , 1989 .

[11]  N. Granichin Udc.  PROCEDURE OF STOCHASTIC APPROXIMATION WITH DISTURBANCES AT THE INPUT , 1992 .

[12]  L. Gerencs.  On Fixed Gain Recursive Estimation Processes , 1996 .

[13]  O. Granichin,et al.  A Randomized Stochastic Approximation Algorithm for Self-Learning , 2005 .

[14]  O. Nelles,et al.  An Introduction to Optimization , 1996, IEEE Antennas and Propagation Magazine.

[15]  James C. Spall,et al.  Developments in stochastic optimization algorithms with gradient approximations based on function measurements , 1994, Proceedings of Winter Simulation Conference.

[16]  Babajide O. Familoni,et al.  A diagonal recurrent neural network-based hybrid direct adaptive SPSA control system , 1999, IEEE Trans. Autom. Control..

[17]  Elizabeth H. Yen,et al.  Weak Sufficient Conditions for Fatou's Lemma and Lebesgue's Dominated Convergence Theorem , 1968 .

[18]  Qing Song,et al.  Dynamic pruning algorithm for multilayer perceptron based neural control systems , 2006, Neurocomputing.

[19]  J. Spall.  Multivariate stochastic approximation using a simultaneous perturbation gradient approximation , 1992 .

[20]  Wansheng Tang,et al.  Simultaneous perturbation stochastic approximation algorithm combined with neural network and fuzzy simulation , 2008 .