Constrained optimization via stochastic approximation with a simultaneous
perturbation gradient approximation
The paper deals with
a projection algorithm for stochastic
approximation using simultaneous perturbation
gradient approximation for optimization
under inequality constraints where no direct gradient of
the loss function is available and
the inequality constraints are given
functions of the optimization parameters.
It is shown that under application of
algorithm, the parameter iterate
converges almost surely to a Kuhn-Tucker point.
The procedure is illustrated by a numerical example.
Keywords: Optimization, Stochastic approximation, SPSA,
Constrained optimization, Inequality constraints, Kuhn-Tucker point.
IMM Technical Report 3/96