David Reshef



I am an MD/PhD computer scientist, broadly interested in the areas of machine learning, statistical inference, and information theory. My work focuses on developing tools for identifying structure in high-dimensional datasets using techniques from these fields.

I completed my MD at Harvard Medical School and the Massachusetts Institute of Technology.

I received my PhD from the department of computer science at the Massachusetts Institute of Technology, where I worked with Josh Tenenbaum and Tommi Jaakkola. I also collaborated closely with Pardis Sabeti and Michael Mitzenmacher.

Previously, I studied statistics at the University of Oxford on a Marshall Scholarship, and computer science at MIT.

Email: dnreshef <at> mit <dot> edu


Selected Publications

Equitability, Interval Estimation, and Statistical Power
Y. Reshef*, D. Reshef*, P. Sabeti**, M. Mitzenmacher**
(* co-first authors; ** co-last authors)

Statistical Science, 2020.
[Abstract] [Manuscript]







An Empirical Study of the Maximal and Total Information Coefficients
and Leading Measures of Dependence

D. Reshef*, Y. Reshef*, P. Sabeti**, M. Mitzenmacher**
(* co-first authors; ** co-last authors)

Annals of Applied Statistics, 2018.
[Abstract] [Manuscript]






Learning Optimal Interventions
J. Mueller, D. Reshef, G. Du, T. Jaakkola
Artificial Intelligence and Statistics (AISTATS), 2017.
[Abstract] [Manuscript]




Measuring dependence powerfully and equitably
Y. Reshef*, D. Reshef*, H. Finucane, P. Sabeti**, M. Mitzenmacher**
(* co-first authors; ** co-last authors)

Journal of Machine Learning Research, 2016.
[Abstract] [Manuscript]





Assessing the Perceived Predictability of Functions
E. Schulz, J. Tenenbaum, D. Reshef, M. Speekenbrink, S. Gershman

Proceedings of the 37th Annual Conference of the Cognitive Science Society, 2015.
[Manuscript]






Cleaning Up the Record on the Maximal Information Coefficient and Equitability
D. Reshef*, Y. Reshef*, M. Mitzenmacher**, P. Sabeti**
(*,** these authors contributed equally)

Proceedings of the National Academy of Sciences (PNAS), 2014.
[Abstract] [Manuscript]





Equitability Analysis of the Maximal Information Coefficient, with Comparisons
D. Reshef*, Y. Reshef*, M. Mitzenmacher**, P. Sabeti**
(*,** these authors contributed equally)

Preliminary version at:
[arXiv]


Factors Related to Increasing Prevalence of Resistance to Ciprofloxacin and Other Antimicrobial Drugs in Neisseria gonorrhoeae, United States
E. Goldstein , R. Kirkcaldy, D. Reshef, S. Berman, H. Weinstock, P. Sabeti, C. Del Rio, G. Hall, E. Hook, M.Lipsitch

Emerging Infectious Diseases, 2012.
[Manuscript]


Detecting novel associations in large datasets
D. Reshef*, Y. Reshef*, H. Finucane, S. Grossman, G. McVean, P. Turnbaugh, E. Lander,
M. Mitzenmacher**, P. Sabeti** (*, ** these authors contributed equally)

Science, 2011.
[Manuscript] [Accompanying Science Perspective] [Science Podcast] [Project Website (reprints)] [FAQ]



On measures of dependence
D. Reshef

Graduate thesis (University of Oxford), Department of Statistics, advised by Gilean McVean, 2011.




Oseltamivir for treatment and prevention of pandemic influenza A/H1N1 virus infection in households, Milwaukee, 2009
E. Goldstein, B. Cowling, J. O'Hagan, L. Danon, V. Fang, A. Hagy, J. Miller, D. Reshef, J. Robins, P. Biedrzycki, M. Lipsitch

BMC Infectious Diseases, 2010.
[Manuscript]


Development of high-throughput drug screening assay for membrane repair
D. Reshef, E. Gallardo, S. Gibb, L. Glover, J. Landers, I. Illa, R. Brown Jr.

Presented at International Dysferlin Conference, 2007.
[Abstract]




Funding

I'm grateful for prior generous support from:
Marshall Aid Commemoration Commission The Paul and Daisy Soros Fellowship The National Science Foundation