Pith. sign in

REVIEW 4 cited by

Certified Defenses against Adversarial Examples

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1801.09344 v2 pith:I2IOGTGI submitted 2018-01-29 cs.LG

Aditi Raghunathan , Jacob Steinhardt , Percy Liang This is my paper · ORCID
classification cs.LG
keywords adversarialcertificatedefensesnetworktestaccuracyattackattacks
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

While neural networks have achieved high accuracy on standard image classification benchmarks, their accuracy drops to nearly zero in the presence of small adversarial perturbations to test inputs. Defenses based on regularization and adversarial training have been proposed, but often followed by new, stronger attacks that defeat these defenses. Can we somehow end this arms race? In this work, we study this problem for neural networks with one hidden layer. We first propose a method based on a semidefinite relaxation that outputs a certificate that for a given network and test input, no attack can force the error to exceed a certain value. Second, as this certificate is differentiable, we jointly optimize it with the network parameters, providing an adaptive regularizer that encourages robustness against all attacks. On MNIST, our approach produces a network and a certificate that no attack that perturbs each pixel by at most \epsilon = 0.1 can cause more than 35% test error.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

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

  1. Robust Representation Consistency Model via Contrastive Denoising

    cs.CV 2025-01 accept novelty 8.0 of 10

    rRCM, a contrastive denoising pre-training and fine-tuning scheme, gives a single-pass robust classifier that beats diffusion-based defenses on ImageNet and CIFAR-10 while reducing inference cost by up to 85x.

  2. TorchLean: Formalizing Neural Networks in Lean

    cs.MS 2026-02 conditional novelty 7.0 of 10

    A Lean 4 framework gives neural networks one machine-checked semantics shared by training, autodiff, and CROWN-style verification, demonstrated on small robustness, PINN, and controller cases.

  3. Enhancing Adversarial Transferability via Component-Wise Transformation

    cs.CV 2025-01 conditional novelty 4.0 of 10

    A block-wise interpolation and selective rotation attack, CWT, improves adversarial transferability across CNN and transformer models on ImageNet.

  4. Unsupervised dense retrieval with conterfactual contrastive learning

    cs.IR 2024-12 conditional novelty 4.0 of 10

    A Shapley-value-based counterfactual regularization improves dense retrievers' robustness to adversarial attacks and enables key passage extraction without passage-level relevance annotations.

Pith tools