Search Results for author:
Found 2 papers, 1 papers with code
Geometric compression of invariant manifolds in neural nets
We confirm these predictions both for a one-hidden layer FC network trained on the stripe model and for a 16-layers CNN trained on MNIST, for which we also find $\beta_\text{Feature}>\beta_\text{Lazy}$.
How isotropic kernels perform on simple invariants
(ii) Next we consider support-vector binary classification and introduce the stripe model where the data label depends on a single coordinate $y(\underline{x}) = y(x_1)$, corresponding to parallel decision boundaries separating labels of different signs, and consider that there is no margin at these interfaces.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
