Deep Learning Theory Symposium

Program

Machine Learning Research Pod

Location

Calvin Lab auditorium and Zoom

Date

Monday, Dec. 6 – Tuesday, Dec. 7, 2021

Back to calendar

All talk times are in Pacific Time.

9 – 9:10 a.m.

Opening Remarks

9:10 – 10:10 a.m.

Uniform Convergence of Interpolators: Gaussian Width, Norm Bounds, and Benign Overfitting

Frederic Koehler (Simons Institute)

10:10 – 11:10 a.m.

Adaptive Wavelet Distillation from DNNs and Dictionary Learning

Bin Yu (UC Berkeley)

11:10 – 11:30 a.m.

Break

11:30 – 11:45 a.m.

Graphon Neural Networks and the Transferability of Graph Neural Networks

Luiz Chamon (UC Berkeley)

11:45 a.m. – 12 p.m.

Self-Training Converts Weak Learners to Strong Learners in Mixture Models

Spencer Frei (UC Berkeley)

12 – 1 p.m.

Lunch

1 – 1:15 p.m.

Non-Parametric Convergence Rates for Plain Vanilla Stochastic Gradient Descent

Rapha,
ël Berthier

1:15 – 1:30 p.m.

Sharp Matrix Concentration

March Boedihardjo (University of California, Irvine)

1:30 – 3 p.m.

Break

3 – 4 p.m.

Reception

9 – 10 a.m.

Learning Staircases

Emmanuel Abbe (École polytechnique fédérale de Lausanne),
Enric Boix (MIT),
Theodor Misiakiewicz (Stanford University)

10 – 11 a.m.

Exact Asymptotics and Universality for Gradient Flows and Empirical Risk Minimizers

Andrea Montanari (Stanford University)

11 – 11:20 a.m.

Break

11:20 – 11:35 a.m.

The Spectrum of Nonlinear Random Matrices for Ultra-Wide Neural Networks

Yizhe Zhu (University of California, Irvine)

11:35 – 11:50 a.m.

Universality of Neural Networks

Dan Mikulincer (MIT)

11:50 a.m. – 1 p.m.

Lunch

1 – 1:15 p.m.

Is Overfitting Actually Benign? On the Consistency of Interpolating Methods

Preetum Nakkiran (UCSD)

1:15 – 1:30 p.m.

On the Cryptographic Hardness of Learning One-Hidden Layer Neural Networks

Ilias Zadik (MIT)

1:30 – 1:50 p.m.

Break

1:50 – 2:50 p.m.

Representation Costs of Linear Neural Networks: Analysis and Design

Mina Karzand (University of California, Davis)