A linear programming approach to sparse linear regression with quantized\n data

2019/03/17 by V. Cerone, Sophie M. Fosson, Cerone, Vito +3
Engineering · Mathematics · #Advanced Optimization Algorithms Research #FOS: Mathematics #Optimization and Control (math.OC) #Sparse and Compressive Sensing Techniques #Statistical Methods and Inference

paper · pdf · doi:10.48550/arxiv.1903.07156

openalex publication_date 2019/03/17 · openalex created_date 2022/07/24 · openalex updated_date 2026/07/28

Abstract

The sparse linear regression problem is difficult to handle with usual sparse\noptimization models when both predictors and measurements are either quantized\nor represented in low-precision, due to non-convexity. In this paper, we\nprovide a novel linear programming approach, which is effective to tackle this\nproblem. In particular, we prove theoretical guarantees of robustness, and we\npresent numerical results that show improved performance with respect to the\nstate-of-the-art methods.\n

Related

• Sparse linear regression with compressed and low-precision data via\n concave quadratic programming
• Sparse Convex Quantile Regression: A Generalized Benders Decomposition Approach
• Sparse Quantile Regression
• Sparse linear regression from perturbed data
• Sparse and Robust Linear Regression: An Optimization Algorithm and Its Statistical Properties