Low-Dimensional Gradient Helps Out-of-Distribution Detection

Low-Dimensional Gradient Helps Out-of-Distribution Detection

Detecting out-of-distribution (OOD) samples is essential for ensuring the reliability of deep neural networks (DNNs) in real-world scenarios. While previous research has predominantly investigated the disparity between in-distribution (ID) and OOD data through forward information analysis, the discr...

Description complète

Détails bibliographiques
Publié dans:IEEE transactions on pattern analysis and machine intelligence. - 1979. - 46(2024), 12 vom: 12. Dez., Seite 11378-11391
Auteur principal: Wu, Yingwen (Auteur)
Autres auteurs: Li, Tao, Cheng, Xinwen, Yang, Jie, Huang, Xiaolin
Format: Article en ligne
Langue:English
Publié: 2024
Accès à la collection:IEEE transactions on pattern analysis and machine intelligence
Sujets:Journal Article
LEADER 01000caa a22002652c 4500
001 NLM377518336
003 DE-627
005 20250306154448.0
007 cr uuu---uuuuu
008 240913s2024 xx |||||o 00| ||eng c
024 7 |a 10.1109/TPAMI.2024.3459988  |2 doi 
028 5 2 |a pubmed25n1257.xml 
035 |a (DE-627)NLM377518336 
035 |a (NLM)39264792 
040 |a DE-627  |b ger  |c DE-627  |e rakwb 
041 |a eng 
100 1 |a Wu, Yingwen  |e verfasserin  |4 aut 
245 1 0 |a Low-Dimensional Gradient Helps Out-of-Distribution Detection 
264 1 |c 2024 
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 08.11.2024 
500 |a published: Print-Electronic 
500 |a Citation Status PubMed-not-MEDLINE 
520 |a Detecting out-of-distribution (OOD) samples is essential for ensuring the reliability of deep neural networks (DNNs) in real-world scenarios. While previous research has predominantly investigated the disparity between in-distribution (ID) and OOD data through forward information analysis, the discrepancy in parameter gradients during the backward process of DNNs has received insufficient attention. Existing studies on gradient disparities mainly focus on the utilization of gradient norms, neglecting the wealth of information embedded in gradient directions. To bridge this gap, in this paper, we conduct a comprehensive investigation into leveraging the entirety of gradient information for OOD detection. The primary challenge arises from the high dimensionality of gradients due to the large number of network parameters. To solve this problem, we propose performing linear dimension reduction on the gradient using a designated subspace that comprises principal components. This innovative technique enables us to obtain a low-dimensional representation of the gradient with minimal information loss. Subsequently, by integrating the reduced gradient with various existing detection score functions, our approach demonstrates superior performance across a wide range of detection tasks. For instance, on the ImageNet benchmark with ResNet50 model, our method achieves an average reduction of 11.15 % in the false positive rate at 95 % recall (FPR95) compared to the current state-of-the-art approach 
650 4 |a Journal Article 
700 1 |a Li, Tao  |e verfasserin  |4 aut 
700 1 |a Cheng, Xinwen  |e verfasserin  |4 aut 
700 1 |a Yang, Jie  |e verfasserin  |4 aut 
700 1 |a Huang, Xiaolin  |e verfasserin  |4 aut 
773 0 8 |i Enthalten in  |t IEEE transactions on pattern analysis and machine intelligence  |d 1979  |g 46(2024), 12 vom: 12. Dez., Seite 11378-11391  |w (DE-627)NLM098212257  |x 1939-3539  |7 nnas 
773 1 8 |g volume:46  |g year:2024  |g number:12  |g day:12  |g month:12  |g pages:11378-11391 
856 4 0 |u http://dx.doi.org/10.1109/TPAMI.2024.3459988  |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 46  |j 2024  |e 12  |b 12  |c 12  |h 11378-11391