Self-supervised learning converts raw perceptual data such as images to ...
In this paper, we explored the use of deep learning for the prediction o...
We introduce CAN, a simple, efficient and scalable method for self-super...
While large language models (LLMs) like GPT-3 have achieved impressive
r...
Integrating functions on discrete domains into neural networks is key to...
Pre-trained language models derive substantial linguistic and factual
kn...
We propose a novel matrix autoencoder to map functional connectomes from...
We consider the question: how can you sample good negative examples for
...
We propose a novel integrated framework that jointly models complementar...
We propose an integrated deep-generative framework, that jointly models
...
A prominent technique for self-supervised representation learning has be...
We study generalization properties of weakly supervised learning. That i...
Strongly log-concave (SLC) distributions are a rich class of discrete
pr...