EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
3914. Solving Stochastic DC programs with DC constraints
Invited abstract in session MC-41: Stochastic and Deterministic Global Optimization, stream Stochastic and Deterministic Global Optimization.
Monday, 12:30-14:00Room: 97 (building: 306)
Authors (first author is the speaker)
| 1. | Thi My Le Le
|
| Computer Science, University of Lorraine | |
| 2. | Hoai An Le Thi
|
| Computer Science, University of Lorraine | |
| 3. | Van Ngai Huynh
|
| Department of Mathematics, University of Quy Nhon |
Abstract
We consider a class of stochastic difference-of-convex-functions (DC) programs, which are optimization problems where the objective function is the expected value of a stochastic DC function based on a probability distribution, while the constraint functions are DC. There have been many methods developed for convex (nonconvex) stochastic optimization problems without constraints or with convex constraints; however, methods designed for nonconvex and nonsmooth constrained programs are rare. The DC algorithm (DCA) is acknowledged in the literature on deterministic optimization as one of the few efficient algorithms for solving large-scale nonconvex and nonsmooth optimization problems. Using penalty techniques, we transform nonconvex constrained stochastic DC programs into standard stochastic DC programs and introduce novel stochastic DCAs to solve the resulting stochastic DC programs. The convergence analysis of the proposed algorithms is thoroughly investigated, and numerical experiments are performed to evaluate their behaviors.
Keywords
- Stochastic Optimization
- Programming, Constraint
Status: accepted
Back to the list of papers