C
Charles Audet
Researcher at École Polytechnique de Montréal
Publications - 218
Citations - 9106
Charles Audet is an academic researcher from École Polytechnique de Montréal. The author has contributed to research in topics: Optimization problem & Constrained optimization. The author has an hindex of 43, co-authored 211 publications receiving 8000 citations. Previous affiliations of Charles Audet include École Polytechnique & HEC Montréal.
Papers
More filters
Journal ArticleDOI
Mesh Adaptive Direct Search Algorithms for Constrained Optimization
Charles Audet,John E. Dennis +1 more
TL;DR: The main result of this paper is that the general MADS framework is flexible enough to allow the generation of an asymptotically dense set of refining directions along which the Clarke derivatives are nonnegative.
Journal ArticleDOI
Analysis of Generalized Pattern Searches
Charles Audet,John E. Dennis +1 more
TL;DR: A simple convergence analysis is provided that supplies detail about the relation of optimality conditions to objective smoothness properties and to the defining directions for the algorithm, and it gives previous results as corollaries.
Book
Derivative-Free and Blackbox Optimization
Charles Audet,Warren Hare +1 more
TL;DR: DFO algorithms have principally fallen into one of two categories: direct search methods and modelbased methods, and researchers began mixing direct search and model-based methods to create hybrid methods with improved performance.
Journal ArticleDOI
A Pattern Search Filter Method for Nonlinear Programming without Derivatives
Charles Audet,John E. Dennis +1 more
TL;DR: This paper formulates and analyzes a pattern search method for general constrained optimization based on filter methods for step acceptance that preserves the division into SEARCH and local POLL steps, which allows the explicit use of inexpensive surrogates or random search heuristics in the SEARCH step.
Journal ArticleDOI
OrthoMADS: A Deterministic MADS Instance with Orthogonal Directions
TL;DR: A new way of choosing directions for the mesh adaptive direct search (Mads) class of algorithms, where the polling directions are chosen deterministically, ensuring that the results of a given run are repeatable, and that they are orthogonal to each other, which yields convex cones of missed directions at each iteration.