Rahul G. Krishnan
About
I am a senior researcher at Microsoft Research New England. In September 2021, I will be an Assistant Professor of Computer Science and Medicine at the University of Toronto. In 2020, I completed my PhD working with Prof. David Sontag at MIT where I was part of the Clinical ML Group. I received a BASc in Computer Engineering at the University of Toronto and an MS from New York University. During my undergraduate degree, among other projects I worked on building factor graphs for clustering and RNA secondary structure prediction.
My research interests include probablistic inference in deep generative models, causality and applications of machine learning to healthcare.
I will be recruiting students at all levels in Fall 2021 at the University of Toronto. If you are interested in developing new machine learning algorithms and applying them to problems in healthcare, apply to work with me here
Research
-
R. Krishnan, Z. Hussain, D. Sontag
Inductive Biases for Unsupervised, Sequential Models of Disease Progression 2020, In submission -
I. Chen, R. Krishnan, D. Sontag
Learning Disease Subtype and Patient Alignment from Longitudinal Data 2020, In submission -
R. Krishnan, S. Cenci, L. Bourouiba
Characterizing bias in epidemic severity due to onset heterogeneity 2020, In submission -
R. Krishnan, A. Khandelwal, R. Ranganath, D. Sontag
Max-Margin learning with the Bayes factor In UAI 2018 -
E. Lehman, R. Krishnan, X. Zhao, R. Mark, L. Lehman
Representation learning approaches to detect false arrhythmia alarms from ECG dynamics In MLHC 2018 -
D. Liang, R. Krishnan, M. Hoffman, T. Jebara
Variational Autoencoders for Collaborative Filtering In WWW 2018 [Code] -
R. Krishnan, D. Liang, M. Hoffman
On the challenges of learning with inference networks on sparse, high-dimensional data In AISTATS 2018 [Code] -
R. Krishnan, M. Hoffman
Inference and Introspection in Deep Generative Models of Sparse Non-Negative Data
Disney Research Award, Presented at Advances in Approximate Bayesian Inference Workshop at N(eur)IPS 2016
-
R. Krishnan, U. Shalit, D. Sontag
Structured Inference Networks for Nonlinear State Space Models In AAAI 2017 [Code] -
R. Krishnan, U. Shalit, D. Sontag
Deep Kalman Filters arXiv:1511.05121 Preprint, 2015
Advances in Approximate Bayesian Inference & Black Box Inference Workshops at N(eur)IPS 2015 [Code] -
R. Krishnan, S. Lacoste-Julien, D. Sontag
Barrier Frank-Wolfe for Marginal Inference In N(eur)IPS 2015 [Code] -
R. Krishnan, N. Razavian, Y. Choi, S. Nigam, S. Blecker, A. Schmidt, D. Sontag
Early Detection of Diabetes from Health Claims Machine Learning in Healthcare Workshop at N(eur)IPS 2013
Education
- Jan. 2017-: PhD at Massachusetts Institute of Technology
- Sept. 2013 - Dec. 2016: MS at New York University
- Sept. 2008 - May. 2013: BaSc. (Computer Engineering) at The University of Toronto
Work
-
June 2016 - Aug. 2016: Research Intern at Adobe Research
Developed methods for improving inference, learning and interpretability in deep generative models of sparse data
Mentor: Matthew Hoffman
-
Sept. 2011 - Sept. 2012: Intern at IBM Canada
Worked on the Single Instruction Multiple Data (SIMD) team to build and test compiler optimization routines for high performance platforms (POWER7, POWER5 and Blue Gene Q).
-
May 2010 - Aug. 2010 : Intern at Microsoft, Redmond
Intern in the Natural Language Group.
Awards
- 2016: NYU: Henning Biermann Award: For outstanding contribution to education or service
- 2013-2016: NYU: McCracken Fellowship
- 2009, 2011: University of Toronto: Adel S. Sedra Award for the Most Outstanding Student in Computer Engineering
-
2008-2013: University of Toronto International Scholarship,
A full 4-year scholarship for undergraduate education. One of four international students selected to receive this award.
Contact
- Email: rahulgk "at" mit.edu
- Office: G32-424, Stata Center, 32 Vassar St, MA 02139