activity
20172021
most citedAdversary Detection in Neural Networks via Persistent Homology

18 citations · 43 across the 4 of their papers we have counts for

collaborators

6 papers

cs.LG20219 cited

A Unified Paths Perspective for Pruning at Initialization

Thomas Gebhart, Udit Saxena, Paul Schrater

A number of recent approaches have been proposed for pruning neural network parameters at initialization with the goal of reducing the size and computational burden of models while…

cs.LG20201 cited

Sheaf Neural Networks

Jakob Hansen, Thomas Gebhart

We present a generalization of graph convolutional networks by generalizing the diffusion operation underlying this class of graph neural networks. These sheaf neural networks are…

cs.SI2020

The Emergence of Higher-Order Structure in Scientific and Technological Knowledge Networks

Thomas Gebhart, Russell J. Funk

The growth of science and technology is a recombinative process, wherein new discoveries and inventions are built from prior knowledge. Yet relatively little is known about the man…

math.AT201915 cited

Path homologies of deep feedforward networks

Samir Chowdhury, Thomas Gebhart, Steve Huntsman +1

We provide a characterization of two types of directed homology for fully-connected, feedforward neural network architectures. These exact characterizations of the directed homolog…

cs.LG2019

Characterizing the Shape of Activation Space in Deep Neural Networks

Thomas Gebhart, Paul Schrater, Alan Hylton

The representations learned by deep neural networks are difficult to interpret in part due to their large parameter space and the complexities introduced by their multi-layer struc…

cs.LG201718 cited

Adversary Detection in Neural Networks via Persistent Homology

Thomas Gebhart, Paul Schrater

We outline a detection method for adversarial inputs to deep neural networks. By viewing neural network computations as graphs upon which information flows from input space to out-…