• Nonparametric Bayesian Statistics

    Bayesian nonparametrics provides modeling solutions by replacing the finite-dimensional prior distributions of classical Bayesian analysis with infinite-dimensional stochastic processes.


Photo of Caroline Uhler
National Science Foundation | April 19, 2017

Caroline Uhler receives NSF CAREER award

Caroline Uhler awarded this National Science Foundation grant to continue her work in graphical models.

MIT Center for Statistics and Data Science | April 19, 2017

What’s Next in Data Science? MIT Professional Education Course

Learn more about the "Data Science: Data to Insight" online course in this short video, which launches again on May 30, 2017. Registration is n...

Boston Globe | April 06, 2017

What makes this the hottest class at MIT?

SDSC Professor Tommi Jaakkola and Regina Barzilay shares why their Introduction to Machine Learning course is one of the most popular courses on ...


Testing properties of distributions over big domains April 28, 2017

Ronitt Rubinfeld (MIT)

Sebastien Roch (Wisconsin) May 05, 2017

Sebastien Roch (Wisconsin)

Invariance and Causality May 12, 2017

Jonas Peters (Univ. of Copenhagen)