Nonconvex Optimization for Statistical Estimation
Abstract
The project aims to address the urgent need for a comprehensive investigation and development of novel mathematical theories and efficient computational methods to deal with important classes of coupled nonconvex nondifferentiable optimization problems arising from modern statistical estimation problems and beyond. Emphasis is placed on bridging theory, computation, and applications by rigorous mathematics. The research methodology is based on unified formulations of the empirical risk minimization in statistical estimation and planning problems under uncertainty as coupled nonconvex nondifferentiable optimization problems with exploitable structures.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 24, 2023
- Accession Number
- AD1230481
Entities
People
- Jong-shi Pang