Relational Verification Leaps Forward with RABBit

Abstract

We propose RABBit, a Branch-and-Bound-based verifier for verifying relational properties defined over Deep Neural Networks, such as robustness against universal adversarial perturbations (UAP). Existing SOTA complete $L_{\infty}$-robustness verifiers can not reason about dependencies between multiple executions and, as a result, are imprecise for relational verification. In contrast, existing SOTA relational verifiers only apply a single bounding step and do not utilize any branching strategies to refine the obtained bounds, thus producing imprecise results. We develop the first scalable Branch-and-Bound-based relational verifier, RABBit, which efficiently combines branching over multiple executions with cross-executional bound refinement to utilize relational constraints, gaining substantial precision over SOTA baselines on a wide range of datasets and networks. Our code is at https://github.com/uiuc-focal-lab/RABBit.

Cite

Text

Suresh et al. "Relational Verification Leaps Forward with RABBit." Neural Information Processing Systems, 2024. doi:10.52202/079017-3920

Markdown

[Suresh et al. "Relational Verification Leaps Forward with RABBit." Neural Information Processing Systems, 2024.](https://mlanthology.org/neurips/2024/suresh2024neurips-relational/) doi:10.52202/079017-3920

BibTeX

@inproceedings{suresh2024neurips-relational,
  title     = {{Relational Verification Leaps Forward with RABBit}},
  author    = {Suresh, Tarun and Banerjee, Debangshu and Singh, Gagandeep},
  booktitle = {Neural Information Processing Systems},
  year      = {2024},
  doi       = {10.52202/079017-3920},
  url       = {https://mlanthology.org/neurips/2024/suresh2024neurips-relational/}
}