EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Architecture-Preserving Provable Repair of Deep Neural Networks

Zhe Tao, Stephanie Nawas, Jacqueline Mitchell, Aditya V. Thakur

Proceedings of the ACM on Programming Languages · 06.06.2023 · 7 (PLDI) · 443-467

DOI: 10.1145/3591238 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the ACM on Programming Languages, online erschienen am 06.06.2023. Verfasst von Zhe Tao und 3 weiteren Autor:innen. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science Foundation, DOE U.S. Department of Energy.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deep neural networks (DNNs) are becoming increasingly important components of software, and are considered the state-of-the-art solution for a number of problems, such as image recognition. However, DNNs are far from infallible, and incorrect behavior of DNNs can have disastrous real-world consequences. This paper addresses the problem of architecture-preserving V-polytope provable repair of DNNs. A V-polytope defines a convex bounded polytope using its vertex representation. V-polytope provable repair guarantees that the repaired DNN satisfies the given specification on the infinite set of points in the given V-polytope. An architecture-preserving repair only modifies the parameters of the DNN, without modifying its architecture. The repair has the flexibility to modify multiple layers of the DNN, and runs in polynomial time. It supports DNNs with activation functions that have some linear pieces, as well as fully-connected, convolutional, pooling and residual layers. To the best our knowledge, this is the first provable repair approach that has all of these features. We implement our approach in a tool called APRNN. Using MNIST, ImageNet, and ACAS Xu DNNs, we show that it has better efficiency, scalability, and generalization compared to PRDNN and REASSURE, prior provable repair methods that are not architecture preserving.

Förderung

  • National Science Foundation
  • DOE U.S. Department of Energy

Zitieren

APA

Tao, Z., Nawas, S., Mitchell, J., & Thakur, A. V. (2023). Architecture-Preserving Provable Repair of Deep Neural Networks. Proceedings of the ACM on Programming Languages, 7(PLDI), 443-467. https://doi.org/10.1145/3591238

Vancouver

Tao Z, Nawas S, Mitchell J, Thakur AV. Architecture-Preserving Provable Repair of Deep Neural Networks. Proceedings of the ACM on Programming Languages. 2023;7(PLDI):443-467. doi:10.1145/3591238

BibTeX
@article{Tao2023,
  author = {Tao, Zhe and Nawas, Stephanie and Mitchell, Jacqueline and Thakur, Aditya V.},
  title = {Architecture-Preserving Provable Repair of Deep Neural Networks},
  journal = {Proceedings of the ACM on Programming Languages},
  year = {2023},
  volume = {7},
  number = {PLDI},
  pages = {443--467},
  doi = {10.1145/3591238}
}
Kostenlos starten →