A Proximal Gradient Framework for Optimization Under Uncertainty with Applications to Sparse Data Representation
Abstract
This paper investigates the application of proximal gradient methods to optimization problems arising in scenarios with inherent uncertainty, particularly focusing on challenges posed by sparse data. We propose a modified proximal gradient algorithm incorporating a robust regularization term designed to mitigate the effects of noise and missing information. The efficacy of the proposed framework is demonstrated through a case study involving sparse behavioral data, drawing connections to recent advancements in representation learning.
Fields of Study & Subjects
Sparse and Compressive Sensing Techniques
Optimization and Variational Analysis
Stochastic Gradient Optimization Techniques
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 14 March 2026
- Language
- EN
- Document type
- Preprint