-
Nonparametric Bayesian Statistics
Bayesian nonparametrics provides modeling solutions by replacing the finite-dimensional prior distributions of classical Bayesian analysis with infinite-dimensional stochastic processes. -
Statistical and Computational Tradeoffs
Computational limitations of statistical problems have largely been ignored or simply overcome by ad hoc relaxations techniques. -
Four Lectures on Causality
On May 10 and 11, Jonas Peters (University of Copenhagen) gave four lectures on causality. Slides and videos are available.
News
Study: Preschoolers learn from math games — to a point
A study co-authored by Professor Esther Duflo finds games for pre-schoolers can improve conceptual math skills, however, the gains may not carry ...
Investigating the trap of unemployment
PhD student Aicha Ben Dhia studies France’s labor market from the perspective of local job-seekers.
QS ranks MIT the world’s No. 1 university for 2017-18
Ranked at the top for the sixth straight year, the Institute also places first in 12 of 46 disciplines.
Upcoming
Sep
08
Sep
14
Special Stochastics and Statistics Seminar
John Cunningham (Columbia University)
September 14 @ 4:00 pm - 5:00 pm
Sep
15
Sample complexity of population recovery
Yury Polyanskiy (MIT)
September 15 @ 11:00 am
Sep
22
Stochastics and Statistics Seminar
Amir Dembo (Stanford University)
September 22 @ 11:00 am