Download PDFOpen PDF in browser

Minimal Modifications of Deep Neural Networks using Verification

19 pagesPublished: May 27, 2020

Abstract

Deep neural networks (DNNs) are revolutionizing the way complex systems are de- signed, developed and maintained. As part of the life cycle of DNN-based systems, there is often a need to modify a DNN in subtle ways that affect certain aspects of its behav- ior, while leaving other aspects of its behavior unchanged (e.g., if a bug is discovered and needs to be fixed, without altering other functionality). Unfortunately, retraining a DNN is often difficult and expensive, and may produce a new DNN that is quite different from the original. We leverage recent advances in DNN verification and propose a technique for modifying a DNN according to certain requirements, in a way that is provably minimal, does not require any retraining, and is thus less likely to affect other aspects of the DNN’s behavior. Using a proof-of-concept implementation, we demonstrate the usefulness and potential of our approach in addressing two real-world needs: (i) measuring the resilience of DNN watermarking schemes; and (ii) bug repair in already-trained DNNs.

Keyphrases: deep neural networks, deep neural networks modification, neural networks verification, neural networks watermarking, verification

In: Elvira Albert and Laura Kovacs (editors). LPAR23. LPAR-23: 23rd International Conference on Logic for Programming, Artificial Intelligence and Reasoning, vol 73, pages 260-278.

BibTeX entry
@inproceedings{LPAR23:Minimal_Modifications_Deep_Neural,
  author    = {Ben Goldberger and Guy Katz and Yossi Adi and Joseph Keshet},
  title     = {Minimal Modifications of Deep Neural Networks using Verification},
  booktitle = {LPAR23. LPAR-23: 23rd International Conference on Logic for Programming, Artificial Intelligence and Reasoning},
  editor    = {Elvira Albert and Laura Kovacs},
  series    = {EPiC Series in Computing},
  volume    = {73},
  publisher = {EasyChair},
  bibsource = {EasyChair, https://easychair.org},
  issn      = {2398-7340},
  url       = {/publications/paper/CWhF},
  doi       = {10.29007/699q},
  pages     = {260-278},
  year      = {2020}}
Download PDFOpen PDF in browser