Contact Me
charshaw [at] mit [dot] edu
My research lies at the interface of causal inference and machine learning. My recent work develops algorithmic tools for improving the design and analysis of randomized experiments. More broadly, I am interested in the intersection of computation and statistics.
This summer, I will be joining the Statistics Department at Columbia as an Assistant Professor. I am currently a FODSI postdoctoral fellow hosted jointly between MIT and UC Berkeley, where I am advised by Costis Daskalakis and Ben Recht.
I obtained my PhD in Computer Science from Yale, where I was advised by Daniel Spielman and Amin Karbasi. My PhD studies were generously supported by an NSF Graduate Research Fellowship. After my PhD, I spent a semester as a postdoctoral fellow at the Simons Institute for the Theory of Computing to participate in the Causality program. During my PhD, I was an intern at Google Research in NYC. I obtained bachelor degrees from Rice University.
You can view my CV here. I am a person who stutters, which you can learn more about here.
Preprints and Working Papers
A Design-based Riesz Representation Framework for Randomized Experiments
Christopher Harshaw, Fredrik Sävje, Yitan Wang
Preprint. 2022.
Best Paper Award at CML4Impact, NeurIPS 2022 Workshop
Optimized Variance Estimation Under Interference and Complex Experimental Designs
Christopher Harshaw, Joel Middleton, Fredrik Sävje
Preprint. 2021.
Academic Publications
Balancing Covariates in Randomized Experiments Using the Gram–Schmidt Walk
Christopher Harshaw, Fredrik Sävje, Daniel Spielman, Peng Zhang.
Journal of the American Statistical Association (JASA). 2023. [code] [talk]
Clip-OGD: An Experimental Design for Adaptive Neyman Allocation in Sequential Experiments
Jessica Dai, Paula Gradu, Christopher Harshaw
NeurIPS. 2023. Spotlight Presentation. [code][talk]
Design and Analysis of Bipartite Experiments Under a Linear Exposure-Response Model
Christopher Harshaw, Fredrik Sävje, David Eisenstat, Vahab Mirrokni, Jean Pouget-Abadie
Electronic Journal of Statistics. 2023. [code] [talk]
How Do You Want Your Greedy: Simultaneous or Repeated?
Moran Feldman, Christopher Harshaw, Amin Karbasi.
Journal of Machine Learning Research. 2023. [code]
Algorithmic Advances for the Design and Analysis of Randomized Experiments
Christopher Harshaw
PhD Dissertation, Fall 2021
The Power of Subsampling in Submodular Optimization
Christopher Harshaw, Ehsan Kazemi, Moran Feldman, Amin Karbasi.
Mathematics of Operations Research. 2021.
Submodular Maximization Beyond Non-negativity: Guarantees, Fast Algorithms, and Applications
Christopher Harshaw, Moran Feldman, Justin Ward, Amin Karbasi.
ICML 2019. [code][talk]
Projection-Free Online Optimization with Stochastic Gradient: From Convexity to Submodularity
Lin Chen, Christopher Harshaw, Hamed Hassani, Amin Karbasi.
ICML 2018. [code][talk]
Greed is Good: Near-Optimal Submodular Maximization via Greedy Optimization
Moran Feldman, Christopher Harshaw, Amin Karbasi.
COLT 2017.
Graph Prints: Towards a Graph Analytic Method for Network Anomaly Detection
Christoper Harshaw, Robert A. Bridges, Michael D. Iannacone, Joel W. Reed, John R. Goodall.
CISRC, 2016. Won Best Paper Award.
Software
GSWDesign.jl
A Julia package for sampling from the Gram–Schmidt Walk Design. We also have a wrapper for R, available here.
SubmodularGreedy.jl
A Julia package containing fast implementations of greedy-based methods for constrained submodular maximization.
Music
I love to write and play music with friends. Here are some of my favorite projects.
The Clinic (2017-2021) New Haven, CT
A jazz / funk ensemble bringing along odd divisions. [soundcloud]
Steve Cox’s Beard (2012-2016) Houston, TX
Large jazz fusion ensemble living in the pocket. [soundcloud] [spotify]
Solo Work (1994-current)
Experimental electronic composition and acoustic songs. [soundcloud]