Conference Schedule



9:30 AM - 1:00 PM Eastern Daylight Time (EDT)


MSML2020 Daily Conference Schedule

Monday 7/20

9:30    Welcoming Comments

9:35    Weinan E - Towards a Mathematical Understanding of Supervised Learning: what we know and what we don't

10:20   Roberto Car - Boosting ab-initio molecular dynamics with machine learning


11:05  The Slow Deterioration of the Generalization Error of the Random Feature Model.  Chao Ma.

11:20  On the stable recovery of deep structured linear networks under sparsity constraints. François Malgouyres.

11:35  Non-Gaussian processes and neural networks at finite widths.  Sho Yaida.

11:50 Gating creates slow modes and controls phase-space complexity in GRUs and LSTMs.  Tankut Can.

12:05 New Potential-Based Bounds for the Geometric-Stopping Version of Prediction with Expert Advice.  Vladimir Kobzar.

12:20 - 12:50 Q&A

Tuesday 7/21

9:30    Stéphane Mallat - Descartes versus Bayes: Harmonic Analysis for High Dimensional Learning and Deep Nets

10:15   Lexing Ying - Solving Inverse Problems with Deep Learning


11:00  Precise asymptotics for phase retrieval and compressed sensing with random generative priors.  Bruno Loureiro.

11:15  Deep Learning Interpretation: Flip Points and Homotopy Methods.  Roozbeh Yousefzadeh.

11:30  Neural network integral representations with the ReLU activation function.  Armenak Petrosyan. 

11:45  DP-LSSGD: A Stochastic Optimization Method to Lift the Utility in Privacy-Preserving ERM.  Bao Wang.

12:00  Geometric Wavelet Scattering Networks on Compact Riemannian Manifolds.  Michael Perlmutter.

12:15  Data-driven Compact Models for Circuit Design and Analysis.  K. Aadithya.

12:30 - 1:00  Q&A

Wednesday 7/22

9:30    George Karniadakis - (PINNs) - Physics Informed Neural Networks: Algorithms, Theory, and Applications

10:15   Stanley Osher - A Machine Learning Framework for Solving High-Dimensional Mean Field Game Problems


11:00  A type of generalization error induced by initialization in deep neural networks. Yaoyu Zhang.

11:15  Quantum Ground States from Reinforcement Learning.  Austen Lamacraft

11:30  NeuPDE: Neural Network-Based Ordinary and Partial Differential Equations for Modeling Time-Dependent Data.  Hayden Schaeffer.

11:45  Deep Fictitious Play for Finding Markovian Nash Equilibrium in Multi-Agent Games.   Ruimeng Hu.

12:00 Large deviations for the perceptron model and consequences for active learning.  Hugo Cui.

12:15  Policy Gradient-based Quantum Approximate Optimization Algorithm.  Jiahao Yao.

12:30 - 1:00  Q&A

Thursday 7/23

9:30    Lenka Zdeborova - The role of data structure in learning shallow neural networks.

10:15   Anna Gilbert - Metric representations: Algorithms and Geometry


11:00  Rademacher Complexity and Spin Glasses: A Hidden link between the Replica and Statistical theories of Learning. Florent Krzakala

11:15  Calibrating Multivariate Lévy Processes with Neural Networks.  Kailai Xu.

11:30  Borrowing From the Future: An Attempt to Address Double Sampling. Yuhua Zhu.

11:45  Deep Domain Decomposition Method: Elliptic Problems.  Xueshuang Xiang.

12:00  Landscape Complexity for the Empirical Risk of Generalized Linear Models.  Antoine Maillard.

12:15 Butterfly-Net2: Simplified Butterfly-Net and Fourier Transform Initialization.  Zhongshu Xu.

12:30 - 1:00 Q&A

Friday 7/24


10:30    Nathan Kutz  - Deep Learning for the Discovery of Coordinates and Dynamics 


11:15  Deep learning Markov and Koopman models with physical constraints.  Hao Wu.

11:30 SelectNet: Learning to Sample from the Wild for Imbalanced Data Training.  Haizhao Yang.

11:45 Robust Training and Initialization of Deep Neural Networks:  An Adaptive Basis Viewpoint. Nathaniel A Trask.

12:00  SchrödingeRNN: Generative Modeling of Raw Audio as a Continuously Observed Quantum State.  Beñat Mencia Uranga.

12:15 - 12:45 Q&A

Online Conference Registration         Powered by SmartChair System