Mladen Kolar
Department Chair and Visiting Professor of Statistics and Data Science
Research Interests
Professor Kolar’s research focuses on developing statistical and machine learning models for scientific discovery from high-dimensional, noisy data. His work combines efficient optimization techniques with theoretically grounded methods to create models that not only predict but also reveal underlying data-generating mechanisms. Unlike black-box models, his approaches prioritize interpretability, making them valuable across domains such as life sciences and social sciences. Professor Kolar’s recent work explores scalable optimization, including stochastic algorithms for constrained problems and distributed methods for federated learning, enabling reliable and interpretable machine learning systems at scale. Email
Prior to joining MBZUAI, Professor Kolar earned his Ph.D. in Machine Learning from Carnegie Mellon University in 2013. He is a professor in the Department of Data Sciences and Operations at the USC Marshall School of Business. Professor Kolar’s research focuses on high-dimensional statistical methods, probabilistic graphical models, and scalable optimization techniques, driven by the goal of uncovering meaningful and scientifically significant structures from observational data. In recognition of his outstanding contributions to these areas, he was awarded the 2024 Junior Leo Breiman Award. He currently serves as an associate editor for the Journal of Machine Learning Research, the Annals of Statistics, the Journal of Computational and Graphical Statistics, and the New England Journal of Statistics in Data Science.
- Ph.D. in Machine Learning from Carnegie Mellon University
- B.Sc. in Computer Engineering from the University of Zagreb
- Junior Leo Breiman Award 2024
- TUM Global Visiting Professor 2021
- J. T. Oden Faculty Fellow 2015
- SIGKDD Dissertation Awards, Honorable mention 2014
- Simons-Berkeley Research Fellow 2013
- Facebook Fellowship 2010 – 2011
- Rector’s Award, University of Zagreb 2006
- Award “Znanost”, Croatian Science Foundation 2006
- “SCIENCE” award for the best undergraduate paper in the field of technical sciences
- Award Josip Lončar 2006
- Given for the best student in the class
- Scholarship “Top Stipendija” 2005
- Prestigious scholarship given to the best 25 students in the country
- ACM Central Europe Programming Contest, 7th place 2002
- Participated in 2001, 2003
- International Olympiad in Informatics, bronze medal 2001
- Adaptive Client Sampling in Federated Learning via Online Learning with Bandit Feedback B. Zhao, L. Wang, Z. Liu, Z. Zhang, J. Zhou, C. Chen, and M. Kolar Journal of Machine Learning Research (2025).
- Statistical Inference for Networks of High-Dimensional Point Processes X. Wang, M. Kolar, and A. Shojaie Journal of the American Statistical Association (2024).
- Fully Stochastic Trust-Region Sequential Quadratic Programming for Equality-Constrained Optimization Problems Y. Fang, S. Na, M. W. Mahoney, and M. Kolar SIAM Journal on Optimization (2024).
- An adaptive stochastic sequential quadratic programming with differentiable exact augmented lagrangians S. Na, M. Anitescu, and M. Kolar Mathematical Programming (2022).
- Two-sample inference for high-dimensional Markov networks B. Kim, S. Liu, and M. Kolar Journal of the Royal Statistical Society. Series B. (2021).
- Estimating differential latent variable graphical models with applications to brain connectivity S. Na, M. Kolar, and O. Koyejo Biometrika (2021).
- Convergent Policy Optimization for Safe Reinforcement Learning M. Yu, Z. Yang, M. Kolar, and Z. Wang Advances in Neural Information Processing Systems (NeurIPS) (2019).
- ROCKET: Robust confidence intervals via Kendall’s tau for transelliptical graphical models R. F. Barber and M. Kolar Annals of Statistics (2018).
- Efficient Distributed Learning with Sparsity J. Wang, M. Kolar, N. Srebro, and T. Zhang International Conference on Machine Learning (ICML) (2017).
- Estimating time-varying networks M. Kolar, L. Song, A. Ahmed, and E. P. Xing Annals of Applied Statistics (2010).