PeopleofSuperintelligence

Researcher

John Duchi

Also known as John C. Duchi

John Duchi is a Stanford researcher in statistical learning and optimization whose 2011 paper with Elad Hazan and Yoram Singer introduced AdaGrad, a stochastic gradient method with per-parameter learning rates. Trained in computer science at Stanford, he completed a Berkeley PhD in EECS in 2014 under Michael Jordan and Martin Wainwright, with a dissertation on trading computation, communication and privacy against statistical performance. Back at Stanford as a faculty member, he won a NeurIPS 2017 best paper award with his doctoral student Hongseok Namkoong for Variance-based Regularization with Convex Objectives. His honours include a 2017 Sloan Research Fellowship, a 2019 ONR Young Investigator Award and the inaugural SIAM Activity Group on Optimization Early Career Prize in 2020. He is an associate professor of Statistics and Electrical Engineering, and in March 2026 a project he leads on the data AI agents need to reflect honest human preferences won funding from the AI Security Institute's Alignment Project.

Born
1984
Badges

Years link to their sources.

Timeline

  1. 2007–2008

    Computer science at Stanford

    He earns a BS in computer science at Stanford University in 2007 and an MS in computer science there in 2008.

  2. 2010–2011

    The AdaGrad algorithm

    He wins the Best Student Paper Award at the International Conference on Machine Learning in 2010. In 2011, he publishes Adaptive Subgradient Methods for Online Learning and Stochastic Optimization with Elad Hazan and Yoram Singer in the Journal of Machine Learning Research. The paper introduces AdaGrad, a stochastic gradient method with per-parameter learning rates that draws on the geometry of data seen in earlier iterations.

  3. 2012–2014

    A Berkeley PhD

    At the University of California, Berkeley, he earns an MA in statistics in 2012 and a PhD in EECS in 2014, advised by Michael Jordan and Martin Wainwright. His dissertation, Multiple Optimality Guarantees in Statistical Learning, studies how estimators can trade between computation, communication and privacy while keeping statistical performance. Berkeley gives him its C.V. Ramamoorthy Distinguished Research Award in 2014.

  4. 2015–2017

    Assistant professor at Stanford

    He joins Stanford as an assistant professor of Statistics and Electrical Engineering, with a courtesy appointment in Computer Science. He receives the Okawa Foundation Award and an honorable mention for the ACM Doctoral Dissertation Award in 2015, and he is named a 2017 Sloan Research Fellow in Mathematics.

  5. 2017

    A NeurIPS best paper

    With his doctoral student Hongseok Namkoong, he wins a NeurIPS 2017 best paper award for Variance-based Regularization with Convex Objectives. The paper builds a convex surrogate for variance from distributionally robust optimization, so the estimator balances bias and variance and improves test performance over standard empirical risk minimization.

  6. 2019

    An ONR Young Investigator Award

    He receives a 2019 Young Investigator Award from the Office of Naval Research and spends spring 2019 as a visiting scientist in the Simons Institute's Data Privacy: Foundations and Applications program. In July 2019, he publishes Statistics and the Oncoming AI Revolution with Emmanuel Candès and Chiara Sabatti in the Harvard Data Science Review.

  7. 2020

    A SIAM optimization prize

    He receives the inaugural SIAM Activity Group on Optimization Early Career Prize in 2020 for contributions to convex, nonconvex and stochastic optimization and to the statistical foundations of optimization methods for data science. Established in 2018 and given every three years, the prize recognizes an early career researcher for contributions in the six calendar years before the award.

  8. As of October 2026

    Associate professor at Stanford

    He is an associate professor of Statistics and Electrical Engineering at Stanford, and by courtesy of Computer Science. In March 2026, (Mis)aligned Preferences, a project he leads with Ben Van Roy and Ramesh Johari, wins funding from the AI Security Institute's Alignment Project to study what data AI agents need to reflect honest human preferences. In spring 2026, he is a visiting scientist and program organizer for the Simons Institute's Federated and Collaborative Learning program.

Career

  1. Since 2015Stanford UniversityAssistant Professor
  2. 2019–2026University of California, BerkeleyVisiting Scientist, Visiting Scientist and Program Organizer

Education

  1. 2014University of California, BerkeleyPhD
  2. 2012University of California, BerkeleyMA in statistics
  3. 2008Stanford UniversityMS
  4. 2007Stanford UniversityBS

From the biography.

  1. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other universityIncludes John Duchi. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other university. We compare doctorate and 5+ year employment links, career timing and current workplaces for 187 major-prize and 1,111 best-paper winners.
  2. 20% of people at companies who shape AI also hold a university role, led by Google and Stanford UniversityIncludes Stanford University. We estimate that 539 of the 6,766 people with a known current organization (8%) hold both a university role and a company role, including 20% of people at companies and 13% of people at universities.
  3. Google had the most people shaping AI every year from 2014 to 2025Includes Stanford University. Our estimates of which organizations employed the most people shaping AI each year from 1955 to 2025, tracing the shift from university labs to companies.
  4. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionIncludes Stanford University. We estimate where the people shaping AI studied and where their careers led. Massachusetts Institute of Technology leads with 467 former students, including 291 recorded doctorates.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of John Duchi yet. Send us a free-licensed photo.

  1. John Duchi's Profile | Stanford Profiles profiles.stanford.edu. Checked 20 October 2026.
  2. Adaptive Subgradient Methods for Online Learning and Stochastic Optimization jmlr.org. Checked 20 October 2026.
  3. Stochastic gradient descent - Wikipedia en.wikipedia.org. Checked 20 October 2026.
  4. Multiple Optimality Guarantees in Statistical Learning | EECS at UC Berkeley www2.eecs.berkeley.edu. Checked 20 October 2026.
  5. John Duchi - The Mathematics Genealogy Project mathgenealogy.com. Checked 20 October 2026.
  6. 2017 Sloan Research Fellowship winners include two IPAM affiliates - IPAM ipam.ucla.edu. Checked 20 October 2026.
  7. Awards neurips.cc. Checked 20 October 2026.
  8. Variance-based Regularization with Convex Objectives papers.nips.cc. Checked 20 October 2026.
  9. John Duchi receives 2019 ONR Young Investigator Award | Department of Statistics statistics.stanford.edu. Checked 20 October 2026.
  10. John Duchi simons.berkeley.edu. Checked 20 October 2026.
  11. John Duchi · PubPub hdsr.mitpress.mit.edu. Checked 20 October 2026.
  12. Inaugural SIAG/OPT Early Career Prize goes to John Duchi | Department of Statistics statistics.stanford.edu. Checked 20 October 2026.
  13. John Duchi wins SIAM Activity Group on Optimization Early Career Prize | Stanford Electrical Engineering ee.stanford.edu. Checked 20 October 2026.
  14. John Duchi web.stanford.edu. Checked 20 October 2026.
  15. MS&E faculty among AI Alignment Project awardees | Management Science and Engineering msande.stanford.edu. Checked 20 October 2026.