The key component in forecasting demand and consumption of resources in a supply
network is an accurate prediction of real-valued time series. Indeed, both service …

Gradient descent finds a global minimum in training deep neural networks despite the
objective function being non-convex. The current paper proves gradient descent achieves …

A longstanding goal in deep learning research has been to precisely characterize training
and generalization. However, the often complex loss landscapes of neural networks have …

It has long been known that a single-layer fully-connected neural network with an iid prior
over its parameters is equivalent to a Gaussian process (GP), in the limit of infinite network …

This textbook establishes a theoretical framework for understanding deep learning models
of practical relevance. With an approach that borrows from theoretical physics, Roberts and …

Modern generative machine learning models are able to create realistic outputs far beyond
their training data, such as photorealistic artwork, accurate protein structures or …

Batch normalization (BN) is a technique to normalize activations in intermediate layers of
deep neural networks. Its tendency to improve accuracy and speed up training have …

Plasticity, the ability of a neural network to quickly change its predictions in response to new
information, is essential for the adaptability and robustness of deep reinforcement learning …

In deep learning theory, the covariance matrix of the representations serves as aproxy to
examine the network's trainability. Motivated by the success of Transform-ers, we study the …

During the past five years the Bayesian deep learning community has developed
increasingly accurate and efficient approximate inference procedures that allow for …