Title :
Newton´s method for optimization with probabilistic estimation
Author :
Lockman, D. ; Mukai, H.
Author_Institution :
Dept. of Syst. Sci. & Math., Washington Univ., St. Louis, MO, USA
Abstract :
Optimization problems are sometimes characterized by an objective function that cannot be directly evaluated but must be estimated, for example by sampling or Monte Carlo simulation. We propose a stabilized Newton´s method aimed at solving a certain class of such problems. Convergence of the proposed algorithm to stationary points is shown, under suitable assumptions
Keywords :
Newton method; convergence of numerical methods; estimation theory; optimisation; probability; convergence; objective function; probabilistic estimation; stabilized Newton method; stochastic optimization; Convergence; Costs; H infinity control; Newton method; Optimal control; Optimization methods; Sampling methods; Stochastic processes; Uncertainty;
Conference_Titel :
Decision and Control, 1996., Proceedings of the 35th IEEE Conference on
Conference_Location :
Kobe
Print_ISBN :
0-7803-3590-2
DOI :
10.1109/CDC.1996.577248
