This year’s annual conference on learning theory (COLT ’15) will take place in beautiful Paris, July 3-6 2015, right before the International Conference of Machine Learning (ICML) in neighboring Lille.

As the leading annual venue for theoretical machine learning we welcome your submissions in all areas broadly related to learning. Please see the call for papers in the main section of the COLT website for submission instructions and guidelines.

As a significant novelty, this year we will allow for two publication tracks. Papers that are accepted to COLT and pass all review stages can publish in the regular conference proceedings. Alternatively, authors of accepted papers will have the option to publish a 1-page abstract in the proceedings and a full conference version in a dedicated ArXiv repository. The latter option was coordinated with the “Annals of Statistics” journal copyright policy to allow for simultaneous publication in both venues.

We welcome your participation in COLT!

Peter Grunwald and Elad Hazan,
COLT 2015 program chairs