About Me
I am a co-founding partner of Interpretable AI, which delivers interpretable methods and solutions for machine learning and artificial intelligence. I obtained my PhD at the Operations Research Center at MIT advised by Dimitris Bertsimas. My primary research interest is the intersection of modern optimization with problems of statistics and machine learning, with the goal of developing new methods that significantly improve upon the state of the art, valuing both performance and interpretability.
Alongside my research, I am a contributor to various open-source projects. I was the lead developer of the OpenSolver add-in, which enables open-source optimization within Microsoft Excel (230,000+ downloads) and Google Sheets (13,000 weekly active users). I have also contributed to the JuliaOpt suite of packages that enable optimization in the Julia programming language.
I have previously worked as an intern at Google, and I obtained my undergraduate degree (B.E. (Hons)) in Engineering Science from the University of Auckland in New Zealand.
PhD Thesis
-
"Optimal Trees for Prediction and Prescription".
(link)
(preprint)
Journal Articles
-
D. Bertsimas, J. Dunn.
"Optimal Classification Trees".
Machine Learning, 2017. (link) (preprint)
- Winner of the MIT ORC Best Student Paper award, 2016
-
D. Bertsimas, J. Dunn, C. Pawlowski, Y. Zhuo.
"Robust Classification".
INFORMS Journal on Optimization, 2018.
(link)
-
D. Bertsimas, J. Dunn, G. Velmahos, H. Kaafarani.
Surgical Risk is Not Linear: Derivation and Validation of a Novel, User-Friendly, and Machine-Learning-based Predictive, OpTimal-Trees-in-Emergency-Surgery, Risk (POTTER) Calculator.
Annals of Surgery, 2018. (link)
-
D. Bertsimas, J. Dunn, C. Pawlowski, J. Silberholz, A. Weinstein, Y. Zhuo, E. Chen, A. Elfiky.
An Applied Informatics Decision Support Tool for Mortality Predictions in Cancer Patients.
JCO Clinical Cancer Informatics, 2018. (link)
-
D. Bertsimas, J. Dunn, N. Mundru.
Optimal Prescriptive Trees.
INFORMS Journal on Optimization, 2019. (link) (preprint)
-
D. Bertsimas, J. Dunn, E. Gibson, A. Orfanoudaki.
Optimal Survival Trees.
Under review.
-
D. Bertsimas, J. Dunn, T. Trikalinos, Y. Wang.
Improved Triaging of Diagnostic Computed Tomography for Children After Head Trauma.
JAMA Pediatrics. 2019. (link)
-
D. Bertsimas, J. Dunn, M. Li, Y. Zhuo, C. Estrada, C. Nelson, H. Wang.
Targeted workup after initial febrile Urinary Tract Infection: Using a novel machine learning model to identify children most likely to benefit from VCUG.
Journal of Urology. 2019. (link)
Talks
Upcoming
- "Optimal Classification and Regression Trees."INFORMS, Oct 2019
Past
- "Optimal Classification and Regression Trees."INFORMS Analytics, Apr 2019
- "Interpretable AI." MIT ILP Japan Conference, Jan 2019
- "Interpretable AI." MIT ILP R&D Conference, Nov 2018
- "Optimal Prescriptive Trees." INFORMS, Nov 2018
- "Optimal Classification and Regression Trees."Computational and Methodological Statistics, Dec 2017
- "Machine Learning in Surgery and Cancer." MIT Sloan: Innovating Health Systems, Digital Health Transformations, Nov 2017
- "Estimating Risk Of Morbity And Mortality After Emergency Surgery With Machine Learning." INFORMS, Nov 2017
- "Optimal Regression Trees." INFORMS, Nov 2017
- "Personalized Medicine for Traumatic Brain Injury." INFORMS Healthcare, July 2017
- "Optimal Classification Trees." University of Auckland, Dec 2016
- "Optimal Classification Trees." INFORMS, Nov 2016
- "Optimal Trees." MIT ORC Seminar Series, Sep 2016
- "Optimal Trees and Robust Classification." University of Auckland, Jan 2016
- "Optimal Trees." INFORMS, Nov 2015