Fall 2019

Sep 6	Tengyuan Liang (University of Chicago) GANs, Optimal Transport, and Implicit Density Estimation	E18-304
Sep 20	Samory Kpotufe (Columbia) Some New Insights On Transfer Learning	E18-304
Sep 27	Adam Klivans (UT Austin) Frontiers of Efficient Neural-Network Learnability	E18-304
Oct 11	Christopher Moore (Santa Fe Institute) The Planted Matching Problem	E18-304
Oct 18	Stanislav Minsker (USC) Towards Robust Statistical Learning Theory	E18-304
Oct 25	Maria-Pia Victoria-Feser (University of Geneva) Accurate Simulation-Based Parametric Inference in High Dimensional Settings	E18-304
Nov 8	Yudong Chen (Cornell) SDP Relaxation for Learning Discrete Structures: Optimal Rates, Hidden Integrality, and Semirandom Robustness	E18-304
Nov 15	Lenka Zdeborova (Institut de Physique Théorique, CNRS) Understanding Machine Learning with Statistical Physics	E18-304
Nov 22	Tamara Broderick (MIT) Automated Data Summarization for Scalability in Bayesian Inference	E18-304
Dec 6	Simon Tavaré (Columbia) Inferring the Evolutionary History of Tumors	E18-304

Feb 1	Andrea Montanari (Stanford) Optimization of the Sherrington-Kirkpatrick Hamiltonian	32-141
Feb 8	Polina Golland (MIT CSAIL) Medical Image Imputation	E18-304
Feb 15	Zhou Fan (Yale) TAP free energy, spin glasses, and variational inference	E18-304
Feb 22	Nike Sun (MIT) Capacity lower bound for the Ising perceptron	E18-304
Mar 1	Eric Kolaczyk (Boston University) Why Aren’t Network Statistics Accompanied By Uncertainty Statements?	E18-304
Mar 8	Aditya Guntuboyina (UC Berkeley) Univariate total variation denoising, trend filtering and multivariate Hardy-Krause variation denoising	E18-304
Mar 15	Alex Belloni (Duke University) Subvector Inference in Partially Identified Models with Many Moment Inequalities	E18-304
Mar 22	Eliran Subag (New York University) Optimization of random polynomials on the sphere in the full-RSB regime	E18-304
Apr 12	Aaditya Ramdas (Carnegie Mellon University) Exponential line-crossing inequalities	E18-304
Apr 19	Dylan Foster (MIT) Logistic Regression: The Importance of Being Improper	E18-304
Apr 26	Chao Gao (University of Chicago) Robust Estimation: Optimal Rates, Computation and Adaptation	E18-304
May 3	Tracy Ke (Harvard) Optimal Adaptivity of Signed-Polygon Statistics for Network Testing (Tracy Ke, Harvard University)	E18-304
May 10	Will Perkins (University of Illinois at Chicago) Counting and sampling at low temperatures	E18-304

Seminars