Prior knowledge-guided attention in self-supervised vision transformers

K Miao, A Gokul, R Singh, S Petryk, J Gonzalez… - arxiv preprint arxiv …, 2022 - arxiv.org
Recent trends in self-supervised representation learning have focused on removing
inductive biases from training pipelines. However, inductive biases can be useful in settings …

[PDF][PDF] Knowledge-Guided Self-Supervised Vision Transformers for Medical Imaging

K Miao, A Gokul, S Petryk, R Singh, K Keutzer… - 2022 - digicoll.lib.berkeley.edu
Recent trends in self-supervised representation learning have focused on removing
inductive biases from the training process. However, inductive biases can be useful in …

[PDF][PDF] Object-Level Representation Learning for Natural and Medical Images

A Gokul - 2022 - digicoll.lib.berkeley.edu
Object perception and recognition is a fundamental part of visual scene understanding.
Humans, even as young as infants, possess a remarkable ability to perceive and represent …