pith. sign in

arxiv: 1903.08778 · v2 · pith:YAEUSAGYnew · submitted 2019-03-20 · 💻 cs.LG · cs.CR· stat.ML

Provable Certificates for Adversarial Examples: Fitting a Ball in the Union of Polytopes

classification 💻 cs.LG cs.CRstat.ML
keywords algorithmballcomputingnetworksneuralpointwiserobustnessinput
0
0 comments X
read the original abstract

We propose a novel method for computing exact pointwise robustness of deep neural networks for all convex $\ell_p$ norms. Our algorithm, GeoCert, finds the largest $\ell_p$ ball centered at an input point $x_0$, within which the output class of a given neural network with ReLU nonlinearities remains unchanged. We relate the problem of computing pointwise robustness of these networks to that of computing the maximum norm ball with a fixed center that can be contained in a non-convex polytope. This is a challenging problem in general, however we show that there exists an efficient algorithm to compute this for polyhedral complices. Further we show that piecewise linear neural networks partition the input space into a polyhedral complex. Our algorithm has the ability to almost immediately output a nontrivial lower bound to the pointwise robustness which is iteratively improved until it ultimately becomes tight. We empirically show that our approach generates distance lower bounds that are tighter compared to prior work, under moderate time constraints.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. On ANN-enhanced positive invariance for nonlinear flat systems

    eess.SY 2026-04 unverdicted novelty 6.0

    A ReLU neural network approximates the distorted constraint set in flat systems to derive a positively invariant set for control synthesis.

  2. An ANN-Enhanced Approach for Flatness-Based Constrained Control of Nonlinear Systems

    eess.SY 2025-03 unverdicted novelty 5.0

    ReLU neural networks approximate transformed constraints in flat systems as unions of polytopes, enabling mixed-integer programming for guaranteed constraint satisfaction in CLF-based and MPC designs for nonlinear systems.