top of page

Scalable Algorithms for Data-Driven Prediction 

This stream of work focuses on developing exact and approximate algorithms for NP-hard problems in machine learning and statistics that are applicable to modern data sizes, including matrix completion and sparse regression. 


Branch-and-Price for Prescriptive Contagion Analytics (with Jacquillat, A., Wang, K., Rame, M.). In Preparation.  


Slowly Varying Regression Under Sparsity (with Bertsimas, D., Diaglakis, Lami, O. S). arXiv preprint arXiv:2102.10773 (2021). Major Revision at Operations Research. 


Stochastic Cutting Planes for Data-Driven Optimization (with Bertsimas, D.).  INFORMS Journal of Computing 34(5), 2400-2409. 


Fast Exact Matrix Completion: A Unified Optimization Framework for Matrix Completion (with Bertsimas, D.). Journal of Machine Learning Research, 21(231), 1-43 (2020).  


Interpretable Matrix Completion: A Discrete Optimization Approach (with Bertsimas, D.). arXiv preprint arXiv:1812.06647 (2020). Submitted to INFORMS Journal of Computing. 


Scalable Holistic Linear Regression (with Bertsimas, D.). Operations Research Letters (2020).  

bottom of page