Rate My Professor Stephen Wright

Wright's research specializes in numerical optimization, focusing on theory, algorithms, implementations, and applications of continuous optimization, particularly in machine learning, data science, and compressed sensing. He is co-author of textbooks including Numerical Optimization, second edition (Springer, 2006, with Jorge Nocedal; more than 51,000 citations), Primal-Dual Interior-Point Methods (SIAM, 1997), Optimization for Data Analysis (Cambridge University Press, 2022, with Ben Recht), and Linear Programming with MATLAB (SIAM, 2007, with Michael Ferris and Olvi Mangasarian). Key papers include "Hogwild!: A Lock-Free Approach to Parallelizing Stochastic Gradient Descent" (NeurIPS, 2011; NeurIPS Test of Time Award), "Coordinate Descent Algorithms" (Mathematical Programming, 2015), and "Gradient Projection for Sparse Reconstruction" (IEEE Journal of Selected Topics in Signal Processing, 2008). He has coauthored widely used software for linear and quadratic programming and compressed sensing. Honors include election to the National Academy of Engineering (2024), George B. Dantzig Prize (SIAM and Mathematical Optimization Society, 2024), Khachiyan Prize (INFORMS Optimization Society, 2020), SIAM Fellowship, Hilldale Professorship (2024-2026), chair of the Mathematical Optimization Society (2007-2010), and SIAM Board of Trustees membership (2005-2014).