1:00-2:00pm in ANGU 241 (Henry Angus).

Title: Optimization for statistical learning with low dimensional structure: regularity and conditioning

https://www.lijunding.net/

3:00pm-4:00pm in Mathx 1100 (Math Annex). Tea at 2:15pm.

Title: Semidefinite programming in data science: strict complementarity and storage optimal algorithms

https://www.lijunding.net/

3:00pm-4:00pm in ESB 2012 (Earth Sciences Building)

Title: Statistical and Computational Phenomena in Deep Learning

https://www.stat.berkeley.edu/~frei/

1:00-2:00pm in ANGU 241 (Henry Angus)

Title: Implicit regularization and benign overfitting for neural networks in high dimensions

https://www.stat.berkeley.edu/~frei/

PhD student, Oxford University

https://pascalegourdeau.github.io/index.html

Assistant Professor, Concordia University

https://sites.google.com/view/paglia

Associate Professor, Michigan State University

https://users.math.msu.edu/users/iwenmark/

Stochastic Bin Packing with Time-Varying Item Sizes

http://www.cs.cmu.edu/~weinaw/

Structured bandits meets classification: the thresholded linear bandits problem

http://web.uvic.ca/~nmehta/

Log-concave Density Estimation in Undirected Graphical Models

Neural Tangent Kernels

Title: Noisy Sorting Capacity

Abstract: We study the sorting problem in the presence of noise. Unlike the common approach in the literature which aims to minimize the number of pairwise comparisons m to achieve a given desired error probability, our goal is to characterize the maximal ratio (n log n)/m such that the ordering of the elements can be estimated with a vanishing error probability asymptotically.

Theoretical bounds for low-rank projection, and the matrix Bernstein inequality