Канал: MSML21

Reduced Order Modeling using Shallow ReLU Networks with Grassmann Layers

Reduced Order Modeling using Shallow ReLU Networks with Grassmann Layers

Average-Case Integrality Gap for Non-Negative Principal Component Analysis

Average-Case Integrality Gap for Non-Negative Principal Component Analysis

Hessian-Aided Random Perturbation (HARP) Using Noisy Zeroth-Order Queries

Hessian-Aided Random Perturbation (HARP) Using Noisy Zeroth-Order Queries

Some observations on partial differential equations in Barron and multi-layer spaces

Some observations on partial differential equations in Barron and multi-layer spaces

Construction of optimal spectral methods in phase retrieval

Construction of optimal spectral methods in phase retrieval

Adversarial Robustness of Stabilized Neural ODE Might be from Obfuscated Gradients

Adversarial Robustness of Stabilized Neural ODE Might be from Obfuscated Gradients

Dynamic Algorithms for Online Multiple Testing

Dynamic Algorithms for Online Multiple Testing

Solvable Model for Inheriting the Regularization through Knowledge Distillation

Solvable Model for Inheriting the Regularization through Knowledge Distillation

Deep Generative Learning via Euler Particle Transport

Deep Generative Learning via Euler Particle Transport

Spectral Geometric Matrix Completion

Spectral Geometric Matrix Completion

Orientation-Preserving Vectorized Distance Between Curves

Orientation-Preserving Vectorized Distance Between Curves

Generalization and Memorization: The Bias Potential Model

Generalization and Memorization: The Bias Potential Model

Sharp threshold for alignment of graph databases with Gaussian weights

Sharp threshold for alignment of graph databases with Gaussian weights

Implicit form neural network for learning scalar hyperbolic conservation laws

Implicit form neural network for learning scalar hyperbolic conservation laws

A Qualitative Study of the Dynamic Behavior for Adaptive Gradient Algorithms

A Qualitative Study of the Dynamic Behavior for Adaptive Gradient Algorithms

Certifying Robustness of Graph Laplacian-Based Semi-Supervised Learning

Certifying Robustness of Graph Laplacian-Based Semi-Supervised Learning

BEAR: Sketching BFGS Algorithm for Ultra-High Dimensional Feature Selection in Sublinear Memory

BEAR: Sketching BFGS Algorithm for Ultra-High Dimensional Feature Selection in Sublinear Memory

Numerical Calabi-Yau metrics from holomorphic networks

Numerical Calabi-Yau metrics from holomorphic networks

Solving Bayesian Inverse Problems via Variational Autoencoders

Solving Bayesian Inverse Problems via Variational Autoencoders

Borrowing From the Future: Addressing Double Sampling in Model-free Control

Borrowing From the Future: Addressing Double Sampling in Model-free Control

On the emergence of tetrahedral symmetry in the final and penultimate layers (...)

On the emergence of tetrahedral symmetry in the final and penultimate layers (...)

A Data Driven Method for Computing Quasipotentials

A Data Driven Method for Computing Quasipotentials

Interpretable and Learnable Super-Resolution Time-Frequency Representation

Interpretable and Learnable Super-Resolution Time-Frequency Representation

DNNs Are Effective At Learning High-Dimensional Hilbert-Valued Functions From Limited Data

DNNs Are Effective At Learning High-Dimensional Hilbert-Valued Functions From Limited Data

Optimal Policies for a Pandemic: A Stochastic Game Approach and a Deep Learning Algorithm

Optimal Policies for a Pandemic: A Stochastic Game Approach and a Deep Learning Algorithm

Analyzing Finite Neural Networks: Can We Trust Neural Tangent Kernel Theory?

Analyzing Finite Neural Networks: Can We Trust Neural Tangent Kernel Theory?

Kernel-Based Smoothness Analysis of Residual Networks

Kernel-Based Smoothness Analysis of Residual Networks

Phase Retrieval with Holography and Untrained Priors:

Phase Retrieval with Holography and Untrained Priors:

Reconstruction of Pairwise Interactions using Energy-Based Models

Reconstruction of Pairwise Interactions using Energy-Based Models

A deep learning method for solving Fokker-Planck equations

A deep learning method for solving Fokker-Planck equations

Deep Autoencoders: From Understanding to Generalization Guarantees

Deep Autoencoders: From Understanding to Generalization Guarantees

The Gaussian equivalence of generative models for learning with shallow neural networks

The Gaussian equivalence of generative models for learning with shallow neural networks

Ground States of Quantum Many Body Lattice Models via Reinforcement Learning

Ground States of Quantum Many Body Lattice Models via Reinforcement Learning

Multilevel Stein variational gradient descent with applications to Bayesian inverse problems

Multilevel Stein variational gradient descent with applications to Bayesian inverse problems

Decentralized Multi-Agents by Imitation of a Centralized Controller

Decentralized Multi-Agents by Imitation of a Centralized Controller

Parameter Estimation with Dense and Convolutional Neural Networks Applied to the FitzHugh-Nagumo ODE

Parameter Estimation with Dense and Convolutional Neural Networks Applied to the FitzHugh-Nagumo ODE

Active learning with importance sampling

Active learning with importance sampling

Noise-Robust End-to-End Quantum Control using Deep Autoregressive Policy Networks

Noise-Robust End-to-End Quantum Control using Deep Autoregressive Policy Networks

T-D learning with nonlinear function approximation: lazy training and mean field regimes

T-D learning with nonlinear function approximation: lazy training and mean field regimes

A semigroup method for high dimensional committor functions based on neural network

A semigroup method for high dimensional committor functions based on neural network

Session 5: Inverse Problems

Session 5: Inverse Problems

Session 1: Optimization and Algorithms

Session 1: Optimization and Algorithms

Session 4: High-dimensional Statistics

Session 4: High-dimensional Statistics

Session 3: Learning Foundations II

Session 3: Learning Foundations II

Session 2: Learning Foundations I

Session 2: Learning Foundations I

Session 7: PDEs and ODEs

Session 7: PDEs and ODEs

Session 6: Reinforcement Learning and Control

Session 6: Reinforcement Learning and Control

Hessian Estimation via Stein's Identity in Black-Box Problems

Hessian Estimation via Stein's Identity in Black-Box Problems

Practical and Fast Momentum-Based Power Methods

Practical and Fast Momentum-Based Power Methods

Session 8: Computational Physics

Session 8: Computational Physics

Kyle Cranmer

Kyle Cranmer

Vanessa Bohm

Vanessa Bohm

George Booth

George Booth

Christoph Ortner

Christoph Ortner

Michael Brenner

Michael Brenner

Gitta Kutyniok

Gitta Kutyniok

Sid Mishra

Sid Mishra

Julia Westermayr

Julia Westermayr

Andrew Stuart

Andrew Stuart

Plenary: Qianxiao Li

Plenary: Qianxiao Li

Plenary: Marc Mezard

Plenary: Marc Mezard