RobustMap : Visual Exploration of DNN Adversarial Robustness in Generative Latent Space

The article presents a novel approach to visualizing adversarial robustness (called robustness below) of deep neural networks (DNNs). Traditional tests only return a value reflecting a DNN's overall robustness across a fixed number of test samples. Unlike them, we use test samples to train a ge...

Description complète

Détails bibliographiques
Publié dans:IEEE transactions on visualization and computer graphics. - 1996. - 31(2025), 9 vom: 01. Aug., Seite 5801-5815
Auteur principal: Li, Jie (Auteur)
Autres auteurs: Kuang, Jielong
Format: Article en ligne
Langue:English
Publié: 2025
Accès à la collection:IEEE transactions on visualization and computer graphics
Sujets:Journal Article
LEADER 01000caa a22002652c 4500
001 NLM378480758
003 DE-627
005 20250801232456.0
007 cr uuu---uuuuu
008 241004s2025 xx |||||o 00| ||eng c
024 7 |a 10.1109/TVCG.2024.3471551  |2 doi 
028 5 2 |a pubmed25n1516.xml 
035 |a (DE-627)NLM378480758 
035 |a (NLM)39361462 
040 |a DE-627  |b ger  |c DE-627  |e rakwb 
041 |a eng 
100 1 |a Li, Jie  |e verfasserin  |4 aut 
245 1 0 |a RobustMap  |b Visual Exploration of DNN Adversarial Robustness in Generative Latent Space 
264 1 |c 2025 
336 |a Text  |b txt  |2 rdacontent 
337 |a ƒaComputermedien  |b c  |2 rdamedia 
338 |a ƒa Online-Ressource  |b cr  |2 rdacarrier 
500 |a Date Revised 31.07.2025 
500 |a published: Print 
500 |a Citation Status PubMed-not-MEDLINE 
520 |a The article presents a novel approach to visualizing adversarial robustness (called robustness below) of deep neural networks (DNNs). Traditional tests only return a value reflecting a DNN's overall robustness across a fixed number of test samples. Unlike them, we use test samples to train a generative model (GM) and render a DNN's robustness distribution over infinite generated samples within the GM's latent space. The approach extends test samples, enabling users to obtain new test samples to improve feature coverage constantly. Moreover, the distribution provides more information about a DNN's robustness, enabling users to understand a DNN's robustness comprehensively. We propose three methods to resolve the challenges of realizing the approach. Specifically, we (1) map a GM's high-dimensional latent space onto a plane with less information loss for visualization, (2) design a network to predict a DNN's robustness on massive samples to speed up the distribution rendering, and (3) develop a system to supports users to explore the distribution from multiple perspectives. Subjective and objective experiment results prove the usability and effectiveness of the approach 
650 4 |a Journal Article 
700 1 |a Kuang, Jielong  |e verfasserin  |4 aut 
773 0 8 |i Enthalten in  |t IEEE transactions on visualization and computer graphics  |d 1996  |g 31(2025), 9 vom: 01. Aug., Seite 5801-5815  |w (DE-627)NLM098269445  |x 1941-0506  |7 nnas 
773 1 8 |g volume:31  |g year:2025  |g number:9  |g day:01  |g month:08  |g pages:5801-5815 
856 4 0 |u http://dx.doi.org/10.1109/TVCG.2024.3471551  |3 Volltext 
912 |a GBV_USEFLAG_A 
912 |a SYSFLAG_A 
912 |a GBV_NLM 
912 |a GBV_ILN_350 
951 |a AR 
952 |d 31  |j 2025  |e 9  |b 01  |c 08  |h 5801-5815