CVLGAug 3, 2018

Ask, Acquire, and Attack: Data-free UAP Generation using Class Impressions

arXiv:1808.01153v10.0094 citations
AI Analysis70

This addresses the challenge of crafting effective adversarial attacks for deep learning models when data is unavailable, which is incremental but improves over existing data-free approaches.

The paper tackles the problem of generating Universal Adversarial Perturbations (UAPs) without access to data by using class impressions to emulate data samples, achieving state-of-the-art success rates in data-free scenarios and performance close to data-driven methods.

Deep learning models are susceptible to input specific noise, called adversarial perturbations. Moreover, there exist input-agnostic noise, called Universal Adversarial Perturbations (UAP) that can affect inference of the models over most input samples. Given a model, there exist broadly two approaches to craft UAPs: (i) data-driven: that require data, and (ii) data-free: that do not require data samples. Data-driven approaches require actual samples from the underlying data distribution and craft UAPs with high success (fooling) rate. However, data-free approaches craft UAPs without utilizing any data samples and therefore result in lesser success rates. In this paper, for data-free scenarios, we propose a novel approach that emulates the effect of data samples with class impressions in order to craft UAPs using data-driven objectives. Class impression for a given pair of category and model is a generic representation (in the input space) of the samples belonging to that category. Further, we present a neural network based generative model that utilizes the acquired class impressions to learn crafting UAPs. Experimental evaluation demonstrates that the learned generative model, (i) readily crafts UAPs via simple feed-forwarding through neural network layers, and (ii) achieves state-of-the-art success rates for data-free scenario and closer to that for data-driven setting without actually utilizing any data samples.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes