Brian Liu

Brian Liu

Welcome to my website! I am a fourth-year PhD candidate in the Operations Research Center at MIT where I am advised by Professor Rahul Mazumder. My research lies in the intersection of statistical machine learning and discrete optimization, and I focus on developing efficient and explainable algorithms for data science and ML. I am also interested in developing interpretable ML methods for healthcare and public policy.

Prior to graduate school, I was a Data and Applied Scientist at Microsoft where I worked on online advertising and I am originally from the San Francisco Bay Area.

Please find my CV here.

Select Honors and Awards

  • Invited to the 2025 ISyE-MS&E-IOE Rising Stars Workshop
  • 2025 American Statistical Association Statistical Computing Section Best Student Paper Winner
  • 2024 INFORMS Data Mining Society Best Student Paper Competition 1st Place

Publications

Under Review

  • B. Liu and R. Mazumder. Randomization Can Reduce Both Bias and Variance: A Case Study in Random Forests, arxiv.org/abs/2402.12668, 2024. R&R at Journal of Machine Learning Research (JMLR).

  • B. Liu and R. Mazumder. Locally Transparent Rule Sets for Explainable Machine Learning, 2025, Submitted to Operations Research.

Published Works

  • B. Liu and R. Mazumder. MOSS: Multi-Objective Optimization for Stable Rule Sets. In ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2025.
    • 2024 INFORMS Data Mining Society Best Student Paper Competition 1st Place.
  • B. Liu and R. Mazumder. FASTopt: An Optimization Framework for Fast Additive Segmentation. In ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2024.
    • 2025 American Statistical Association Statistical Computing Student Paper Competition Winner.

  • B. Liu and R. Mazumder. FIRE: An Optimization Framework for Fast Interpretable Rule Extraction. In ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2023.

  • B. Liu and R. Mazumder. ForestPrune: Compact Depth-Pruned Tree Ensembles. In Proceedings of the 26th International Conference on Artificial Intelligence and Statistics (AISTATS), 2023.

  • B. Liu , M. Xie, and M. Udell. ControlBurn: Feature Selection by Sparse Forests. In ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2021.

  • B. Liu*, Y. Zhang*, S. Henderson, D. Shmoys, P. Frazier. Modeling the risk of in-person instruction during the COVID-19 pandemic, INFORMS Journal of Applied Analytics, 2024.

  • P. Frazier, J. M. Cashore, N. Duan, S. Henderson, A. Janmohamed, B. Liu , D. Shmoys, J. Wan, Y. Zhang. Modeling for COVID-19 College Reopening Decisions: Cornell, A Case Study. Proceedings of the National Academy of Sciences.

Talks

  • MIT Sloan Health Systems Initiative Annual Workshop, October 2024
    • Interpretable Machine Learning Methods for Predicting Telemental Health Outcomes
  • INFORMS Annual Meeting, October 2024
    • An Optimization Framework for Fast Additive Segmentation in Transparent ML
  • ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, August 2024
    • An Optimization Framework for Fast Additive Segmentation in Transparent ML
  • International Symposium on Mathematical Programming, July 2024
    • An Optimization Framework for Fast Additive Segmentation in Transparent ML
  • US Census Bureau Center for Statistical Research and Methodology, July 2024
    • Making Tree Ensembles Interpretable
  • ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, August 2023
    • Fast Interpretable Rule Extraction
  • INFORMS Annual Meeting, October 2022
    • Depth-Pruning Tree Ensembles
  • ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, August 2021
    • Feature Selection with Sparse Forests