51:40

Qin Li - Mean field theory in Inverse Problems: From Bayesian inference to overparametrized networks

49:56

Jeremy Budd - Joint Reconstruction-Segmentation with Graph PDEs

57:39

Rama Cont - Asymptotic Analysis of Deep Residual Networks

51:06

Anna Korba - Kernel Stein Discrepancy Descent

55:44

Robert Nowak - What Kinds of Functions Do Neural Networks Learn?

39:42

Yury Korolev - Approximation properties of two-layer neural networks with values in a Banach space

49:24

Po-Ling Loh - Robust W-GAN-Based Estimation Under Wasserstein Contamination

53:48

Matthias Ehrhardt - Bilevel Learning for Inverse Problems

51:52

Ferdia Sherry - Structure-preserving machine learning for inverse problems

58:27

Arthur Gretton - Generalized Energy-Based Models

01:01:37

Andrew Stuart - Learning Linear Operators

49:37

George Karniadakis - Approximating functions, functionals and operators using DNNs

45:40

Tatiana Bubba - Deep neural networks for inverse problems with pseudodifferential operators

48:34

Eldad Haber - PDE Inspired Graph Neural Networks

43:18

Bao Wang - Advances of momentum in optimization algorithms and neural architecture design

52:42

Petar Veličkovć - Geometric Deep Learning: Grids, Graphs, Groups, Geodesics and Gauges

42:14

Sebastian Kassing - Convergence of Stochastic Gradient Descent for analytic target functions

01:03:08

Kaushik Bhattacharya - Learning based multi-scale modeling

52:52

Nik Nuesken: Stein geometry in machine learning: gradient flows, optimal transport, large deviations

56:03

Wuchen Li - Transport information Bregman divergences

48:10

Nicolas Garcia Trillos - Adversarial Classification, Optimal Transport, and Geometric Flows

53:23

Derek Driggs - Barriers to Deploying Deep Learning Models During the COVID-19 Pandemic

48:20

Boumediene Hamzi - Machine Learning and Dynamical Systems meet in Reproducing Kernel Hilbert Spaces

01:02:38

Bubacarr Bah - Discrete Optimization Methods for Group Model Selection in Compressed Sensing

51:37

Jeff Calder - Random walks and PDEs in graph-based learning

58:50

Boris Hanin - Finite Width, Large Depth Neural Networks as Perturbatively Solvable Models

49:41

Nathaniel Trask - Structure preservation and convergence in scientific machine learning

52:11

Frederic Koehler - Classification Under Misspecification

49:37

Andrea Agazzi - Convergence & optimality of single-layer neural networks for reinforcement learning

57:01

Melanie Weber - Geometric Methods for Machine Learning and Optimization

51:15

Carola Bibiane Schönlieb - Machine Learned Regularization for Solving Inverse Problems

49:45

Ziwei Ji - The dual of the margin: improved analyses and rates for gradient descent’s implicit bias

51:23

Nadia Drenska - A PDE Interpretation of Prediction with Expert Advice

43:20

Felix Voigtlaender - Neural networks for classification problems with boundaries of Barron class

01:08:21

Bamdad Hosseini - Conditional Sampling with Monotone GANs: Generative Models and Inverse Problems

52:48

Zhengdao Chen - A Dynamical Central Limit Theorem for Shallow Neural Networks

59:57

Jonas Latz - Analysis of Stochastic Gradient Descent in Continuous Time

53:14

Yu Bai - How Important is the Train-Validation Split in Meta-Learning?

01:00:48

Ryan Murray - Consistency of Cheeger cuts: Total Variation, Isoperimetry, and Clustering

43:09

Qi Lei - Predicting What You Already Know Helps: Provable Self-Supervised Learning

55:41

Jason Klusowski - Sparse Learning with CART

01:07:06

Franca Hoffmann - Geometric Insights into Spectral Clustering by Graph Laplacian Embeddings

47:49

Alex Safsten - Stability of Accuracy for Deep Neural Network Classifiers

48:43

Holden Lee - Provable Algorithms for Sampling Non-log-concave Distributions

01:05:07

Nadav Cohen - Analyzing Optimization and Generalization in DL via Dynamics of Gradient Descent

01:07:29

Lénaïc Chizat - Analysis of Gradient Descent on Wide Two-Layer ReLU Neural Networks

50:54

Lei Wu - Understanding flow-based models: Representation, landscape and gradient flow

01:02:38

Soledad Villar - Dimensionality reduction and matching datasets

44:57

Yiping Lu - Optimization Of Neural Network: A Continuous Depth Limit Point Of View And Beyond

01:04:04

Lukasz Szpruch - Mean-Field Neural ODEs, Relaxed Control and Generalization Errors

57:04

Aditi Raghunathan - Tradeoffs between Robustness and Accuracy

01:06:13

Stephan Wojtowytsch - Banach spaces for multi-layer networks and connections to mean field training

56:39

Huy Tuan Pham - A general framework for the mean field limit of multilayer neural networks

01:05:00

Konstantinos Spiliopoulos - Mean field limits of neural networks: typical behavior and fluctuations

53:11

Roberto I. Oliveira - A mean-field theory for certain deep neural networks

01:15:10

Anders Hansen - On the foundations of computational mathematics and the potential limits of AI

01:11:07

Eric Vanden-Eijnden - Trainability and accuracy of artificial neural networks

01:12:53

Weinan E - Towards a mathematical understanding of supervised learning