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.
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.
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
