One of the main challenges in modern deep learning is to understand why ...
While Convolutional Neural Networks (CNNs) have long been investigated a...
Lipschitz continuity is a simple yet pivotal functional property of any
...
Transformers have achieved remarkable success in several domains, rangin...
`Double descent' delineates the generalization behaviour of models depen...
The Hessian of a neural network captures parameter interactions through
...
Second-order information, in the form of Hessian- or Inverse-Hessian-vec...
Combining different models is a widely used paradigm in machine learning...
We propose a method to learn unsupervised sentence representations in a
...
We propose a unified framework for building unsupervised representations...