Rohit Bhattacharya

Mere data makes a man —
A and C and T and G.
The alphabet of you —
All from four symbols.
I am only two — 1 and 0.

I am an Assistant Professor of Computer Science at Williams College. I develop statistical and machine learning methods for causal inference, and study the application of these methods to problems in computational genomics and healthcare. My research is generously supported by an NSF CRII grant.

Previously, I completed my BS in Biomedical Engineering and PhD in Computer Science from Johns Hopkins University where I was advised by Ilya Shpitser.

Email: rb17[at]williams.edu

Twitter: r0ntu

News: (scroll to view older news)

Sep 25, 2024: Paper on proximal causal inference with text has been accepted at NeurIPS 2024 -- camera-ready.

May 31, 2024: Paper on causal null hypothesis tests that are robust to both statistical and causal model misspecification has been accepted at UAI 2024 -- camera-ready.

May 15, 2024: Yufeng Wu and David Goetze successfully defended their respective theses on causal inference under interference and missing data methods for federated learning. Congrats Yufeng and David!

Apr 24, 2024: I have been awarded an NSF CRII grant to study and develop robust methods for causal inference.

Oct 16, 2023: New draft on causal null hypothesis tests that are robust to both statistical and causal model misspecification -- arXiv.

Jun 10, 2023: New paper on causal inference with outcome dependent missingness and self-censoring has been accepted at UAI 2023 -- camera-ready.

Jun 10, 2023: Paper on goodness of fit tests for missing data models has been accepted at UAI 2023 -- camera-ready.

May 15, 2023: Jacob Chen and Elijah Tamarchenko successfully defended their respective theses on causal inference with self-censoring outcomes and variable selection methods for causal inference. Congrats Jacob and Elijah!

February 24, 2023: I was interviewed for a Nature Outlook story on causality and artificial intelligence. You can read about it here.

May 15, 2022: Carson Kurtz successfully defended his thesis on causal discovery for multivariate regression chain graphs. Congrats Carson!

May 15, 2022: The paper on designing tests for assumptions of the front-door model has been accepted at UAI 2022 -- camera-ready.

Feb 28, 2022: New draft on designing goodness-of-fit of tests for missing not at random graphical models arXiv.

Feb 28, 2022: New draft on designing tests for assumptions of the front-door model arXiv.

Nov 29, 2021: Revision for our paper on semiparametric estimation of causal effects in causal graphs with unmeasured confounders on arXiv.

Mar 5, 2021: The paper on differentiable causal discovery in the presence of unmeasured confounders has been accepted at AISTATS 2021 -- camera-ready and code.

Jan 22, 2021: I accepted a tenure-track positition as Assistant Professor of Computer Science at Williams College!

Oct 15, 2020: New draft on differentiable causal discovery in the presence of unmeasured confounders on arXiv.

May 1, 2020: Public release of Ananke: a Python software package for causal inference using graphical models.

Apr 16, 2020: Razieh Nabi and I gave a joint talk on semiparametric inference for causal effects in graphical models with hidden variables at the JHU Biostatistics Causal Inference Seminar.

Jan 15, 2020: Razieh Nabi and my introductory course on Causal Inference was featured in the Johns Hopkins Hub magazine!

May 25, 2019: Razieh Nabi and I won the Thomas R. Ten Have award at ACIC, Montréal. This means that we will be giving a joint talk at ACIC next year in Austin, Texas.

May 22, 2019: Presenting a poster on identification in missing data models represented by directed acyclic graphs at ACIC, Montréal.

May 22, 2019: Presenting a poster on causal inference under interference and network uncertainty at ACIC, Montréal.