Methods for evaluating robustness of solutions to constraint problems include: (a) determining one or more inputs for the constraint problem; (b) determining bias, uncertainty, or bias and uncertainty of the one or more inputs; (c) randomly perturbating at least one of the one or more inputs based on the bias, uncertainty, or bias and uncertainty of the one or more inputs to determine one or more modified inputs; (d) generating a solution to the constraint problem based on the one or more modified inputs; (e) storing the solution; (f) repeating (c) through (e) until multiple solutions are stored; and/or (g) determining the robustness of the stored multiple solutions by comparison of the stored multiple solutions to each other. The one or more modified inputs include the randomly perturbated input or inputs and/or the non-perturbated input or inputs.

 
Web www.patentalert.com

< Method and appartus for using B measures to learn balanced relevance functions from expert and user judgments

> Regularized least squares classification or regression with leave-one-out (LOO) error

> Bipedal walking simulation

~ 00589