241. On Constraint Qualifications for MPECs with Applications to Bilevel Hyperparameter Optimization for Machine Learning
Invited abstract in session MB-7: Hyperparameter Optimization for Classification, stream Bilevel and multilevel optimization.
Monday, 10:30-12:30Room: B100/5015
Authors (first author is the speaker)
| 1. | Jiani Li
|
| Beijing Institute of Technology |
Abstract
Constraint qualifications for a Mathematical Program with Equilibrium Constraints (MPEC) are essential in analyzing different stationary points and establishing convergence results. In this paper, we explore various classical MPEC constraint qualifications and analyze connections between them. We subsequently study the behavior of these constraint qualifications in the context of a specific MPEC derived from the bilevel hyperparameter optimization (BHO) for L1-loss support vector classification. In particular, for such an MPEC, we provide a full characterization of the well-known
MPEC linear constraint qualification, therefore, establishing conditions under which it holds or fails for our BHO for support vector machines.
Keywords
- Complementarity and variational problems
- Linear and nonlinear optimization
Status: accepted
Back to the list of papers