Search Results for author:
Found 2 papers, 0 papers with code
A Scalable Walsh-Hadamard Regularizer to Overcome the Low-degree Spectral Bias of Neural Networks
We show how this spectral bias towards low-degree frequencies can in fact hurt the neural network's generalization on real-world datasets.
Contrastive Learning of Single-Cell Phenotypic Representations for Treatment Classification
Learning robust representations to discriminate cell phenotypes based on microscopy images is important for drug discovery.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
