My research interests are broadly in interpretable machine learning, data-driven analytics, and optimization.
Many machine learning problems reduce to solving an optimization problem. My research focuses on developing methods to solve these challenging optimizations problems that arise in many different machine learning settings. In particular, my optimization work has revolved around solving non-convex and non-smooth problems.
Another focus of my research work has been developing interpretable machine learning methods that facilitate the decision-making process for domain experts. Specifically, in my recent work, we developed a hybrid model where an interpretable model is constructed to compete and collaborate with any pre-trained black-box model, gaining model transparency at no or low cost of predictive performance.
I am also interested in data visualization and data-driven analytics with applications in health, sports, and other fields.
Interpretable Machine Learning
H. Rafique, T. Wang, Q. Lin, and A. Singhani. Transparency Promotion with Model-Agnostic Linear Competitors. Proceedings of the Thirty Seven International Conference on Machine Learning (ICML), 2020.
H. Rafique, T. Wang, and Q. Lin. Model-Agnostic Linear Competitors - When Interpretable Models Compete and Collaborate with Black-box Models, Third INFORMS Workshop on Data Science, INFORMS College on Artificial Intelligence, 2019. Best Paper Award Runner-Up
M. Liu , H. Rafique, Q. Lin, and T. Yang. Provable Non-Convex Min-Max Optimization, Smooth game optimization and machine learning workshop, Neurips 2018
M. Liu , H. Rafique, Q. Lin, and T. Yang. First-order Convergence Theory for Weakly-Convex-Weakly-Concave Min-max Problems . Journal of Machine Learning, 2021.
H. Rafique, M. Liu, Q. Lin, and T. Yang. Weakly-convex–concave min–max optimization: provable algorithms and applications in machine learning, Optimization Methods and Software, 2021.
M. Imran, H. Rafique, A. Khan, and T. Malik. Model of Bi-mode Transmission Dynamics of Hepatitis C with Optimal Control, Theory in Biosciences, 2014.
Manuscripts In Progress
H. Rafique and Q. Lin. Second-Order Trust-Region Method for Non-Convex Min-Max Problems.