no code implementations • 7 Jun 2023 • Tomojit Ghosh, Michael Kirby, Karim Karimov
In the first step, we solve the linear Centroid-Encoder, a convex optimization problem over a matrix $A$.
no code implementations • 7 Jun 2023 • Tomojit Ghosh, Michael Kirby
During training, we update class centroids by taking the Hadamard product of the centroids and weights of the sparse layer, thus ignoring the irrelevant features from the target.
no code implementations • 7 Jun 2023 • Tomojit Ghosh, Michael Kirby
SLCE works by mapping the samples of a class to its class centroid using a linear transformation.
no code implementations • 30 Jan 2022 • Tomojit Ghosh, Michael Kirby
The resulting algorithm, Sparse Centroid-Encoder (SCE), extracts discriminatory features in groups using a sparsity inducing $\ell_1$-norm while mapping a point to its class centroid.
no code implementations • 29 Sep 2021 • Tomojit Ghosh, Michael Kirby
We develop a sparse optimization problem for the determination of the total set of features that discriminate two or more classes.
no code implementations • 27 Feb 2020 • Tomojit Ghosh, Michael Kirby
The Centroid-Encoder (CE) method is similar to the autoencoder but incorporates label information to keep objects of a class close together in the reduced visualization space.