Ankur Moitra

Moitra grew up in Niskayuna, New York, surrounded by a family of computer scientists; he recalls deciding early on that computer science was “definitely not cool” before discovering computer science and mathematics on his own and falling in love with them.⁠[1]

In 2007 he received a B.S. in electrical and computer engineering from Cornell UniversityCornell UniversityCornell University is a private research university in Ithaca, New York, and a member of the Ivy League. It is New York State's federal land-grant university, a distinction unique within the Ivy League. Ezra Cornell and Andrew Dickson White founded the university in 1865, and it opened in 1868. In addition to its Ithaca campus, Cornell operates Weill Cornell Medicine and Cornell Tech in New York City and a medical college in Qatar. It enrolled 26,793 students in fall 2024. Its president is Michael I. Kotlikoff.Open the full entry, where he was the valedictorian.⁠[2]

He received his M.S. and PhD in computer science from the Massachusetts Institute of Technology in 2009 and 2011, advised by Tom Leighton; his master’s and doctoral dissertations won the William A. Martin Award and the George M. Sprowls Award for best thesis.⁠[3]

He was supported by a Fannie and John Hertz Foundation Fellowship; his doctoral thesis introduced vertex sparsification: for a huge network in which only a few terminals need to communicate, a much smaller network approximates the relevant communication properties of the original and can serve as a proxy for running algorithms.⁠[4]

After his PhD he was an NSF Computing and Innovation Fellow at the Institute for Advanced Study and also a senior postdoc in the computer science department at Princeton UniversityPrinceton UniversityPrinceton University is a private university and Ivy League member in Princeton, New Jersey. Chartered on October 22, 1746 as the College of New Jersey, it moved to Princeton in 1756 and adopted its present name in 1896. Undergraduate education leads to A.B. and B.S.E. degrees and emphasizes small-group discussion, independent research and residential college life. Graduate education spans the humanities, social sciences, natural sciences and engineering. As of October 2026, Christopher L. Eisgruber was president. Its 2026–27 operating budget was $3.407 billion, including $342 million for undergraduate aid.Open the full entry.⁠[3]

On April 23, 2010, he and Gregory Valiant submitted “Settling the Polynomial Learnability of Mixtures of Gaussians” to arXiv, giving an algorithm that estimates the parameters of a mixture of multivariate Gaussians with running time and data polynomial in the dimension and inverse accuracy, and proving that exponential dependence on the number of Gaussians is necessary.⁠[5]

In October 2010 he and his advisor Tom Leighton published “Some Results on Greedy Embeddings in Metric Spaces” in Discrete & Computational Geometry, resolving the conjecture of Papadimitriou and Ratajczak that every 3-connected planar graph admits a greedy embedding into the Euclidean plane, a question arising from geographic routing in wireless sensor networks.⁠[6]

On November 3, 2011, he, Sanjeev Arora, Rong Ge and Ravi Kannan submitted “Computing a Nonnegative Matrix Factorization -- Provably” to arXiv; for the NP-complete nonnegative matrix factorization problem, it gave a polynomial-time algorithm for constant rank with a matching hardness result, and a potentially practical algorithm under a separability condition.⁠[7]

On April 9, 2012, he, Arora and Ge submitted “Learning Topic Models - Going beyond SVD,” using nonnegative matrix factorization to give, under a separability assumption, the first polynomial-time algorithm for learning topic models that neither assumes one topic per document nor recovers only the span of the topic vectors.⁠[8]

In 2013 he returned to MIT as an assistant professor of applied mathematics and a principal investigator in the Computer Science and Artificial Intelligence Laboratory (CSAIL).⁠[9]

On his homepage he states that a major goal of his work is to give algorithms with provable guarantees for problems in machine learning; he is a member of MIT’s Theory of Computation group and an associate editor of the Journal of the ACMAustralian Community MediaAustralian Community Media (ACM) is an Australian regional and rural news publishing group with more than 100 brands, including The Canberra Times; in 2019 Nine sold it to a company controlled by interests associated with Antony Catalano and Thorney Investment Group.Open the full entry and Communications of the ACM.⁠[10]

In 2015 he received an NSF CAREER Award and the MIT Mathematics Department’s Edmund F. Kelly Research Award.⁠[11]

On April 21, 2016, he, Ilias Diakonikolas, Gautam Kamath, Daniel Kane, Jerry Li and Alistair Stewart submitted “Robust Estimators in High Dimensions without the Computational Intractability” to arXiv, studying high-dimensional distribution learning when an adversary may arbitrarily corrupt a fraction of the samples and giving the first efficient algorithms with dimension-independent error for several fundamental classes of distributions.⁠[12]

In October 2016, then the Rockwell International Career Development Associate Professor, he was named one of 18 David and Lucile Packard Fellows for the year, receiving a five-year unrestricted research grant totaling $875,000 for research on algorithmic aspects of machine learning; department head Tomasz Mrowka called him “the dream colleague.”⁠[13]

The same year he also received a Sloan Research Fellowship.⁠[11]

He was promoted to associate professor in July 2017 and received tenure in July 2019.⁠[9]

In 2018 Cambridge University Press published his book Algorithmic Aspects of Machine Learning (151 pages), which bridges theoretical computer science and machine learning and emphasizes the need for flexible, tractable models that capture what makes machine learning easy rather than what makes it hard.⁠[14]

That year he received the MIT School of Science Prize for Excellence in Graduate Teaching and a Young Investigator Award from the Office of Naval Research (ONR).⁠[11]

In fall 2021 he was an organizer of the Simons Institute for the Theory of Computing program “Computational Complexity of Statistical Inference”; in fall 2024 and spring 2025 he took part as a visiting scientist in the institute’s special year on large language models and transformers.⁠[15]

A 2021 CSAIL profile notes that he and his wife have two children; he said he often does his deepest thinking late at night, alone and free of distractions.⁠[1]

In June 2021 MIT announced that he would become director of the Statistics and Data Science Center (SDSC) in the Institute for Data, Systems, and Society effective July 1, 2021, succeeding founding director Devavrat Shah, and would become a full professor that month; the announcement noted that every summer he and David Jerison mentor twenty or more high school and undergraduate research projects.⁠[16]

In July 2021 he was named the Norbert Wiener Professor of Mathematics.⁠[11]

In January 2024 he gave the American Mathematical Society’s John von Neumann Lecture on “Learning From Dynamics”; in fall 2024 he was Chancellor’s Professor in Electrical Engineering and Computer Science at the University of California, BerkeleyUniversity of California, BerkeleyThe University of California, Berkeley (UC Berkeley) is a campus of the University of California system; the University of California was founded in 1868 through a merger of two institutions and moved to Berkeley, California, in 1873.Open the full entry.⁠[17]

While developing a fast algorithm for quantum spin systems at high temperature, he, Ainesh Bakshi, Allen Liu and Ewin Tang unexpectedly proved that in spin systems in thermal equilibrium entanglement vanishes completely above a certain temperature that does not depend on the total number of atoms; Moitra commented, “It’s not just that it’s exponentially small … It’s zero.”⁠[18]

He received the 2026 IEEE Computer Society W. Wallace McDowell Award “for groundbreaking contributions to high-dimensional learning—spanning mixture models, robust statistics, and quantum systems.”⁠[2]

In an interview on the award he said that theory with provable guarantees is about building abstractions that test whether we really understand AI models, describing playing with them as “like an alien technology.”⁠[19]

The 2016 robust estimation paper, presented at FOCS 2016 and published in the SIAM Journal on Computing, volume 48 (2019), received the 2026 Gödel Prize awarded by ACM SIGACT and EATCS; the citation says it resolved a longstanding problem in robust statistics and “helped launch modern algorithmic high-dimensional robust statistics.”⁠[20]

The Gödel Prize has been awarded annually since 1993 and carries a cash prize of US$5,000; as of June 2026 the paper had been cited more than 670 times.⁠[21]

In summer 2026 his chair in the MIT Department of Mathematics changed from the Norbert Wiener Professorship to the Francis Wright Davis Professorship.⁠[17]

On August 3, 2026, MIT announced that Alexander Rakhlin would succeed him as SDSC director, a post he had held since 2021, with Philippe Rigollet serving as interim director in 2024–25; the announcement described him as also associate director of IDSS and a faculty member in Electrical Engineering and Computer Science.⁠[22]

He appeared with Steven BruntonSteven BruntonSteven Brunton is a professor of mechanical engineering at the University of Washington whose research combines machine learning with dynamical systems to model and control complex systems such as fluid flows; with collaborators he introduced the SINDy method for sparse identification of nonlinear dynamics, and he teaches engineering mathematics on YouTube as “eigensteve.”Open the full entry and Alex Kontorovich on episode 503 of the Lex FridmanLex FridmanLex Fridman is a Soviet-born, US-raised computer scientist and podcast host who works as an MIT research scientist on human-AI interaction and autonomous driving. His Lex Fridman Podcast is known for hours-long conversations with guests including heads of state and technology leaders.Open the full entry Podcast.⁠[23]

Personal website: Public link: https://people.csail.mit.edu/moitra/ (opens in a new window)⁠[10]

Massachusetts Institute of Technology: Official website: https://www.mit.edu/ (opens in a new window)⁠[9]

Institute for Advanced Study: Official website: https://www.ias.edu/ (opens in a new window)⁠[4]

Princeton University: Official website: https://www.princeton.edu/ (opens in a new window)⁠[4]

University of California, Berkeley: Official website: https://www.berkeley.edu/ (opens in a new window)⁠[17]