Conference on Learning Theory

The 38th Annual Conference on Learning Theory (COLT 2025) takes place June 30–July 4, 2025 in Lyon, France.

Photo by Dimitry B. on flickr

Important Dates

All dates are in 2025.

February 6 (5pm ET) Submission deadline
March 28 Reviews released
April 4 Initial author response due
April 4–11 Discussion period
May 2 Author notification
June 30–July 4 Conference
Photo by Eric Rothermel on Unsplash

Conference Organization

Program Chairs

Workshops, Tutorials, & Community Events Chairs

Open Problems Chair

Publication Chair

Workshops, Tutorials, & Community Events

These events take place on June 30, 2025.

Workshops

Tutorials

  • How to Make Use of Learning Theory to Learn Efficient ML Models: From PAC-Bayesian Generalization Bounds to (Self-Bounding) Learning Algorithms
    • Emilie Morvant, Paul Viallard
  • Language Generation in the Limit
    • Moses Charikar, Anay Mehrotra, Chirag Pabbaraju, Charlotte Peale, Grigoris Velegkas
  • Efficient Learning with Smoothed Data: Bridging Statistical and Online Learning
    • Adam Block, Abhishek Shetty

Community event

  • Crafting and Communicating a Research Agenda
    • Surbhi Goel, Vatsal Sharan, Jess Sorrell, Lydia Zakynthinou, Tijana Zrnic

Satellite event

  • Sharp and Foundry: On Frugal and Robust Foundations for Machine Learning

Sponsors

We are very grateful to our sponsors whose extraordinary generosity has been critical to the continued success of the conference, particularly by making it possible for students to attend and present their research. For information about sponsorship opportunities, please contact the sponsorship chairs at sponsorships@learningtheory.org.

Contact Information