Abstract
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 structure. We introduce a method for computing persistent homology over the graphical activation structure of neural networks, which provides access to the task-relevant substructures activated throughout the network for a given input. This topological perspective provides unique insights into the distributed representations encoded by neural networks in terms of the shape of their activation structures. We demonstrate the value of this approach by showing an alternative explanation for the existence of adversarial examples. By studying the topology of network activations across multiple architectures and datasets, we find that adversarial perturbations do not add activations that target the semantic structure of the adversarial class as previously hypothesized. Rather, adversarial examples are explainable as alterations to the dominant activation structures induced by the original image, suggesting the class representations learned by deep networks are problematically sparse on the input space.
Original language | English (US) |
---|---|
Title of host publication | Proceedings - 18th IEEE International Conference on Machine Learning and Applications, ICMLA 2019 |
Editors | M. Arif Wani, Taghi M. Khoshgoftaar, Dingding Wang, Huanjing Wang, Naeem Seliya |
Publisher | Institute of Electrical and Electronics Engineers Inc. |
Pages | 1537-1542 |
Number of pages | 6 |
ISBN (Electronic) | 9781728145495 |
DOIs | |
State | Published - Dec 2019 |
Event | 18th IEEE International Conference on Machine Learning and Applications, ICMLA 2019 - Boca Raton, United States Duration: Dec 16 2019 → Dec 19 2019 |
Publication series
Name | Proceedings - 18th IEEE International Conference on Machine Learning and Applications, ICMLA 2019 |
---|
Conference
Conference | 18th IEEE International Conference on Machine Learning and Applications, ICMLA 2019 |
---|---|
Country/Territory | United States |
City | Boca Raton |
Period | 12/16/19 → 12/19/19 |
Bibliographical note
Publisher Copyright:© 2019 IEEE.
Keywords
- Adversarial Examples
- Deep Learning
- Neural Networks
- Persistent Homology
- Topology