# Harnessing the Vulnerability of Latent Layers in Adversarially Trained Models

Source: [Harnessing the Vulnerability of Latent Layers in Adversarially Trained Models](<https://www.sota2.com/research/paper/68fa65f11e209131cea0d431>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T21:38:30.146Z

Authors: Mayank Singh, Abhishek Sinha, Nupur Kumari, Harshitha Machiraju, Balaji Krishnamurthy, Vineeth N Balasubramanian

Published: 2019-05-13T16:44:03.000Z

## Abstract

Neural networks are vulnerable to adversarial attacks -- small visually imperceptible crafted noise which when added to the input drastically changes the output. The most effective method of defending against these adversarial attacks is to use the methodology of adversarial training. We analyze the adversarially trained robust models to study their vulnerability against adversarial attacks at the level of the latent layers. Our analysis reveals that contrary to the input layer which is robust to adversarial attack, the latent layer of these robust models are highly susceptible to adversarial perturbations of small magnitude. Leveraging this information, we introduce a new technique Latent Adversarial Training (LAT) which comprises of fine-tuning the adversarially trained models to ensure the robustness at the feature layers. We also propose Latent Attack (LA), a novel algorithm for construction of adversarial examples. LAT results in minor improvement in test accuracy and leads to a state-of-the-art adversarial accuracy against the universal first-order adversarial PGD attack which is shown for the MNIST, CIFAR-10, CIFAR-100 datasets.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Image Classification on CIFAR-10 (test) (Adversarial Robustness - PGD 100 & Auto Attack)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-adversarial-robustness-pgd-100-and-auto-attack.md>): Image Classification; CIFAR-10 (test); Clean Accuracy: 87.8.

- [Image Classification on CIFAR-10 (test)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test.md>): Image Classification; CIFAR-10 (test); Accuracy: 87.8.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1905.05186>)

[PDF](<https://arxiv.org/pdf/1905.05186.pdf>)