Prof. John A. Smith | MIT Computer Science and Artificial Intelligence Laboratory

About

I am a Professor in the Department of Electrical Engineering and Computer Science at MIT, and a member of the Computer Science and Artificial Intelligence Laboratory (CSAIL). My research focuses on artificial intelligence, machine learning, and their applications to scientific discovery.

Prior to joining MIT, I was an Assistant Professor at Stanford University (2010-2015) and received my Ph.D. from Carnegie Mellon University in 2009. I am a recipient of the MacArthur Fellowship (2018), NSF CAREER Award (2016), and Sloan Research Fellowship (2017).

Education

Ph.D. in Computer Science, Carnegie Mellon University, 2009
M.S. in Computer Science, University of California, Berkeley, 2005
B.S. in Electrical Engineering and Computer Science, MIT, 2003

Awards and Honors

MacArthur Fellowship, 2018
National Academy of Engineering Member, 2022
AAAI Fellow, 2021
NSF CAREER Award, 2016
Sloan Research Fellowship, 2017

Research

My research group develops new machine learning methods to solve fundamental problems in science and engineering. We are particularly interested in:

Research Areas

Scientific Machine Learning: Developing AI methods that can accelerate scientific discovery in physics, chemistry, and biology
Interpretable AI: Creating machine learning models that provide human-understandable explanations
AI for Climate Science: Applying deep learning to climate modeling and renewable energy optimization
Foundation Models: Building large-scale models that can generalize across multiple scientific domains

Current Projects

DARPA AI for Scientific Discovery (2023-2026)
NSF Expeditions in Machine Learning for Climate Science (2024-2029)
MIT-IBM Watson AI Lab Collaboration on Interpretable AI (2022-2025)

Selected Publications

Scientific Discovery with Generative Models

Smith, J.A., Chen, X., Johnson, R., et al.

Nature, 2025 (in press)

Interpretable Neural Networks for Physical Systems

Smith, J.A., Williams, T., Brown, L.

Proceedings of the National Academy of Sciences, 122(15), 2024

Climate Modeling with Physics-Informed Machine Learning

Lee, H., Smith, J.A., Patel, R., et al.

Science Advances, 10(3), 2024

Foundations of Scientific Machine Learning

Smith, J.A.

Annual Review of Computer Science, 8, 2023

Teaching

Current Courses

6.883: Interpretability and Explainability in Machine Learning (Spring 2026)
6.S897: Machine Learning for Scientific Discovery (Fall 2025)

Past Courses

Introduction to Machine Learning (6.862, Fall 2024)
Advanced Topics in AI (6.864, Spring 2024)
Neural Networks for Physical Systems (6.S979, Fall 2023)

Student Advising

I currently advise 8 Ph.D. students and 2 postdoctoral associates. Prospective students should apply through the MIT EECS graduate program.

Research Group

Current Members

Xia Chen - Postdoctoral Associate (PhD Stanford 2023)
Ryan Johnson - PhD Candidate (MIT 2021-)
Lisa Brown - PhD Candidate (MIT 2022-)
Thomas Williams - PhD Student (MIT 2023-)
Hannah Lee - PhD Student (MIT 2024-)

Alumni

Dr. Rachel Patel - Assistant Professor at UC Berkeley (PhD MIT 2023)
Dr. Michael Zhang - Research Scientist at DeepMind (PhD MIT 2022)
Dr. Sarah Wilson - Research Scientist at OpenAI (PhD MIT 2021)

Contact

Email: smith@mit.edu

Office: 32-G768, Ray and Maria Stata Center

Address:
MIT Computer Science and Artificial Intelligence Laboratory
32 Vassar Street
Cambridge, MA 02139 USA

Administrative Assistant: Jane Doe (jdoe@csail.mit.edu, 617-253-1234)

Prospective students: Please apply through the MIT EECS graduate program.