About
I am an assistant professor of information systems at Carnegie Mellon University's Heinz College, and an affiliated faculty member of the Machine Learning Department. I work on machine learning for healthcare and for sustainable development. I am also a co-founder and advisor for the agriculture technology startup CoolCrop, which provides farmers in India with refrigeration and marketing analytics.
My research is on forecasting, such as predicting how long a patient will stay in a hospital, or how produce prices will change in a week at over a thousand Indian markets. To produce forecasts, I typically use nonparametric methods that make very few assumptions on the underlying data. Since these methods inform interventions that can be costly and affect people's well-being, ensuring that predictions are reliable is essential. To this end, in addition to developing nonparametric predictors, I also produce theory to understand when and why they work, and I identify forecast evidence to help practitioners make decisions.
Postdoc opening: I'm currently looking for a postdoc generally interested in nonparametric methods for prediction and causal inference, especially in time series settings. Email me directly if you're interested.
Prospective PhD students: If you are interested in working with me but have not already been accepted into a PhD program at Carnegie Mellon University, then apply to a CMU PhD program first instead of contacting me (PhD admissions are handled at the department or college level depending on the program, and are not done by me individually, so I cannot grant you admission or say whether you will get accepted). Specifically within CMU's Heinz College, the PhD programs most relevant to my work are the joint PhD program in Machine Learning and Public Policy and the PhD program in Information Systems and Management. If you are already a CMU PhD student and would like to work with me, please contact me directly.
Email: georgechen [at symbol] cmu.edu
Office: HBH 2216 (the west wing of Hamburg Hall, second floor)
Teaching schedule (academic year 2019-2020): 95-865 "Unstructured Data Analytics" (minis 2 and 3), 94-775 "Unstructured Data Analytics for Policy" (mini 3)
Current Projects
- Debiasing matrix completion with entries missing not at random (with Wei Ma): we're presenting some preliminary results at NeurIPS 2019, with code available here; we received the Best Paper award (theoretical track) at the INFORMS Data Mining and Decision Analytics Workshop 2019
- Forecasting hospital ICU length of stay (with Lexie Li and Jeremy Weiss; previous contributors: Amanda Coston and Ren Zuo)
- Nonparametric time series classification to predict outcomes of heart attack patients who go into a coma, where we take advantage of EEG data (with Junyan Pu and Jonathan Elmer)
- Looking at what features in educational videos may be predictive of them getting watched, and interpeting the features using SHAP (with Mi Zhou, Pedro Ferreira, and Michael Smith): this is Mi's job market paper — she's already presented this work at INFORMS/CIST 2019 and will be presenting at WISE 2019
- Understanding how reputation influences persuasiveness over time (with Emaad Manzoor and Dokyun Lee): Emaad has presented preliminary results at the ISMS Marketing Science Conference 2019 and INFORMS/CIST 2019
- Providing small and marginal farmers in rural India with access to cost-effective reliable refrigeration, and predictive marketing analytics (with an Indian ag tech startup CoolCrop; previous contributors: Wei Ma and Fred Yang): check out these news articles (here and here) or our ICTD 2019 paper
Pre-historic
Before joining Carnegie Mellon, I finished my Ph.D. in Electrical Engineering and Computer Science at MIT, advised by Polina Golland and Devavrat Shah. My thesis was on nonparametric machine learning methods for forecasting viral news, recommending products to people, and finding human organs in medical images. I also worked on satellite image analysis to help bring electricity to rural India, and modeled brain activation patterns evoked by reading sentences. Between grad school and becoming faculty, I helped develop the recommendation engine at a predictive analytics startup Celect (which has since been acquired by Nike) and then was a teaching postdoc in MIT's Digital Learning Lab, where I was the primary instructor and course developer for a new edX course on computational probability and inference.
I enjoy teaching and pondering the future of education! I have previously taught at MIT, UC Berkeley, and in Jerusalem at a program MEET that brings together Israeli and Palestinian high school students. As a grad student, I served on the Task Force on the Future of MIT Education, and my time as a teaching postdoc was all about better understanding the digital learning space.
Last updated 11/1/2019.