Slides

Samuel Schoenholz - Priors for Deep Infinite Networks

Sho Yaida - Fluctuation-Dissipation Relation for Stochastic Gradient Descent

Levent Sagun - Over-parametrization in neural networks: observations and a definition

Ekin Cubuk - Generalization under noise, adversarial examples, and data augmentation

Dmitri Chklovskii - Neuroscience-based machine learning

Dmitry Krotov - Unsupervised Learning on a Sphere

Jack Hidary & Stefan Leichenauer - Neural network classifiers on NISQ-regime processors

Dar Gilboa - Signal propagation and gradient stability in the LSTM and GRU

Felix Draxler - Essentially No Barriers In Neural Network Energy Landscape

Vladimir Kirilin - Gradient and hessian properties in logistic regression

Roger Grosse - Scalable Natural Gradient Training of Neural Networks

Zohar Ringel - The role of a layer in supervised learning

Jeffrey Pennington - Statistics of Random Neural Networks

Sam McCandlish - An Empirical Model of Large-Batch Training

Katherine Quinn - Visualizing Probabilities: Intensive Principal Component Analysis

Aristide Baratin - On the Spectral Bias of Neural Networks

Joshua Batson - Noise2Self: Blind Denoising with Self-Supervision

Joan Bruna - Geometric Deep Learning in a NonCommutative World

Alexander Alemi - TherML: A thermodynamics of machine learning

Jonathan Kadmon - High-dimensional manifold classification by multilayered perceptrons

Gavin Hartnett - Replica Symmetry Breaking in Bipartite Spin Glasses and Neural Networks

Yasaman Bahri - Wide, Deep Neural Networks as Gaussian Processes

Behnam Neyshabur - The Role of Over Parametrization in the Generalization of Neural Networks

Samuel Smith - What can stochastic differential equations tell us about stochastic gradient descent and natural gradient descent?

Paul Ginsparg - Rise of the machines: deep learning from Backgammon to Skynet

Uros Seljak - EL2O and MPM: alternatives to Variational Inference and MAP/MLE

Andrea Montanari - A mean field view of the landscape of two layers neural networks

Maxime Gabella - Introduction to Topological Data Analysis

Alex Cole - Local Deformations and Global Features in Persistent Homology, Machine Learning, and Physics

Jesse Thaler - Collision Course: Particle Physics as a Machine Learning Testbed

David Spergel - ML for sub-grid physics

Jascha Sohl-Dickstein - Parameter Estimation in Intractable Probabilistic Models by Minimum Probability Flow Learning

Adam Scherlis - The Goldilocks zone: Towards better understanding of neural network loss

Anders Andreassen - JUNIPR: a Framework for Unsupervised Machine Learning in Particle Physics

David Shih - Searching for New Physics with Deep Autoencoders

Lechao Xiao - Why ConvNets generalize better than fully-connected networks? An explanation via a mean field approach

Peter Eckersley - Impossibility and Uncertainty Theorems in AI value alignment (or why your AI should not have a utility function)

Masoud Mohseni - Approximate optimization and inference with tensor networks

Michael Douglas - Machine Learning by a String (Theorist)

Daniel Park - Optimal SGD Hyperparameters for Fully-connected Networks

Samuel Ocko - Diversity versus Depth in Neural Representations

Miles Stoudenmire - "Wavefunctions" of data: applying tensor network methods from physics to machine learning