Search Results for author:
Found 2 papers, 1 papers with code
Relating Regularization and Generalization through the Intrinsic Dimension of Activations
Given a pair of models with similar training set performance, it is natural to assume that the model that possesses simpler internal representations would exhibit better generalization.
Verifying the Union of Manifolds Hypothesis for Image Data
Assuming that data lies on a single manifold implies intrinsic dimension is identical across the entire data space, and does not allow for subregions of this space to have a different number of factors of variation.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
