6 citations · 12 across the 5 of their papers we have counts for
5 papers
Densely Connected -invariant Deep Neural Networks with Signed Permutation Representations
Devanshu Agrawal, James Ostrowski
We introduce and investigate, for finite groups , -invariant deep neural network (-DNN) architectures with ReLU activation that are densely connected-- i.e., include all p…
A Classification of -invariant Shallow Neural Networks
Devanshu Agrawal, James Ostrowski
When trying to fit a deep neural network (DNN) to a -invariant target function with a group, it only makes sense to constrain the DNN to be -invariant as well. However, t…
A Group-Equivariant Autoencoder for Identifying Spontaneously Broken Symmetries
Devanshu Agrawal, Adrian Del Maestro, Steven Johnston +1
We introduce the group-equivariant autoencoder (GE-autoencoder) -- a deep neural network (DNN) method that locates phase boundaries by determining which symmetries of the Hamiltoni…
Wide Neural Networks with Bottlenecks are Deep Gaussian Processes
Devanshu Agrawal, Theodore Papamarkou, Jacob Hinkle
There has recently been much work on the "wide limit" of neural networks, where Bayesian neural networks (BNNs) are shown to converge to a Gaussian process (GP) as all hidden layer…
Fiber Bundles and Parseval Frames
Devanshu Agrawal, Jeff Knisley
Continuous frames over a Hilbert space have a rich and sophisticated structure that can be represented in the form of a fiber bundle. The fiber bundle structure reveals the central…