Corpus ID: 168170195

A Topology Layer for Machine Learning

@inproceedings{Gabrielsson2020ATL,
  title={A Topology Layer for Machine Learning},
  author={Rickard Br{\"u}el Gabrielsson and Bradley J. Nelson and Anjan Dwaraknath and P. Skraba and L. Guibas and G. Carlsson},
  booktitle={AISTATS},
  year={2020}
}
Topology applied to real world data using persistent homology has started to find applications within machine learning, including deep learning. We present a differentiable topology layer that computes persistent homology based on level set filtrations and distance-bases filtrations. We present three novel applications: the topological layer can (i) serve as a regularizer directly on data or the weights of machine learning models, (ii) construct a loss on the output of a deep generative network… Expand
37 Citations
A Topological Framework for Deep Learning
  • 2
  • PDF
Efficient Topological Layer based on Persistent Landscapes
  • 4
Topological Deep Learning: Classification Neural Networks
  • PDF
Domain Adaptation by Topology Regularization
  • Highly Influenced
  • PDF
Supervised Machine Learning From A Topological Perspective
Topology and Neural Networks
PLLay: Efficient Topological Layer based on Persistent Landscapes
  • 3
  • PDF
A Topological Loss Function for Deep-Learning based Image Segmentation using Persistent Homology
  • 19
  • PDF
Regularization of Persistent Homology Gradient Computation
  • 2
  • PDF
TopoGAN: A Topology-Aware Generative Adversarial Network
  • 8
  • PDF
...
1
2
3
4
...

References

SHOWING 1-10 OF 42 REFERENCES
Adversary Detection in Neural Networks via Persistent Homology
  • 11
  • PDF
Deep Learning with Topological Signatures
  • 107
  • PDF
Neural Persistence: A Complexity Measure for Deep Neural Networks Using Algebraic Topology
  • 31
  • PDF
Characterizing the Shape of Activation Space in Deep Neural Networks
  • 6
Applying Topological Persistence in Convolutional Neural Network for Music Audio Signals
  • 24
  • PDF
A Topological Regularizer for Classifiers via Persistent Homology
  • 36
  • PDF
Adversarial Examples Target Topological Holes in Deep Networks
  • 2
  • PDF
On Characterizing the Capacity of Neural Networks using Algebraic Topology
  • 46
  • PDF
...
1
2
3
4
5
...