---
_id: '18211'
abstract:
- lang: eng
text: 'Deep neural networks are known to be vulnerable to malicious perturbations.
Current methods for improving adversarial robustness make use of either implicit
or explicit regularization, with the latter is usually based on adversarial training.
Randomized smoothing, the averaging of the classifier outputs over a random distribution
centered in the sample, has been shown to guarantee a classifier’s performance
subject to bounded perturbations of the input. In this work, we study the application
of randomized smoothing to improve performance on unperturbed data and increase
robustness to adversarial attacks. We propose to combine smoothing along with
adversarial training and randomization approaches, and find that doing so significantly
improves the resilience compared to the baseline. We examine our method’s performance
on common whitebox (FGSM, PGD) and black-box (transferable attack and NAttack)
attacks on CIFAR-10 and CIFAR-100, and determine that for a low number of iterations,
smoothing provides a significant performance boost that persists even for perturbations
with a high attack norm, . For example, under a PGD-10 attack on CIFAR-10 using
Wide-ResNet28-4, we achieve 60.3% accuracy for infinity norm ∞ = 8/255 and 13.1%
accuracy for ∞ = 35/255 – outperforming previous art by 3% and 6%, respectively.
We achieve nearly twice the accuracy on ∞ = 35/255 and even more so for perturbations
with higher infinity norm. A reference implementation of the proposed method is
provided. '
article_processing_charge: No
article_type: original
author:
- first_name: Yaniv
full_name: Nemcovsky, Yaniv
last_name: Nemcovsky
- first_name: Evgenii
full_name: Zheltonozhskii, Evgenii
last_name: Zheltonozhskii
- first_name: Chaim
full_name: Baskin, Chaim
last_name: Baskin
- first_name: Brian
full_name: Chmiel, Brian
last_name: Chmiel
- first_name: Alexander
full_name: Bronstein, Alexander
id: 58f3726e-7cba-11ef-ad8b-e6e8cb3904e6
last_name: Bronstein
orcid: 0000-0001-9699-8730
- first_name: Avi
full_name: Mendelson, Avi
last_name: Mendelson
citation:
ama: Nemcovsky Y, Zheltonozhskii E, Baskin C, Chmiel B, Bronstein AM, Mendelson
A. Adversarial robustness via noise injection in smoothed models. Applied Intelligence.
2022;53(8):9483-9498. doi:10.1007/s10489-022-03423-5
apa: Nemcovsky, Y., Zheltonozhskii, E., Baskin, C., Chmiel, B., Bronstein, A. M.,
& Mendelson, A. (2022). Adversarial robustness via noise injection in smoothed
models. Applied Intelligence. Springer Nature. https://doi.org/10.1007/s10489-022-03423-5
chicago: Nemcovsky, Yaniv, Evgenii Zheltonozhskii, Chaim Baskin, Brian Chmiel, Alex
M. Bronstein, and Avi Mendelson. “Adversarial Robustness via Noise Injection in
Smoothed Models.” Applied Intelligence. Springer Nature, 2022. https://doi.org/10.1007/s10489-022-03423-5.
ieee: Y. Nemcovsky, E. Zheltonozhskii, C. Baskin, B. Chmiel, A. M. Bronstein, and
A. Mendelson, “Adversarial robustness via noise injection in smoothed models,”
Applied Intelligence, vol. 53, no. 8. Springer Nature, pp. 9483–9498, 2022.
ista: Nemcovsky Y, Zheltonozhskii E, Baskin C, Chmiel B, Bronstein AM, Mendelson
A. 2022. Adversarial robustness via noise injection in smoothed models. Applied
Intelligence. 53(8), 9483–9498.
mla: Nemcovsky, Yaniv, et al. “Adversarial Robustness via Noise Injection in Smoothed
Models.” Applied Intelligence, vol. 53, no. 8, Springer Nature, 2022, pp.
9483–98, doi:10.1007/s10489-022-03423-5.
short: Y. Nemcovsky, E. Zheltonozhskii, C. Baskin, B. Chmiel, A.M. Bronstein, A.
Mendelson, Applied Intelligence 53 (2022) 9483–9498.
date_created: 2024-10-08T12:47:53Z
date_published: 2022-08-09T00:00:00Z
date_updated: 2024-10-09T11:04:54Z
day: '09'
doi: 10.1007/s10489-022-03423-5
extern: '1'
fulldoi: https://doi.org/10.1007/s10489-022-03423-5
intvolume: ' 53'
issue: '8'
language:
- iso: eng
month: '08'
oa_version: None
page: 9483-9498
publication: Applied Intelligence
publication_identifier:
eissn:
- 1573-7497
issn:
- 0924-669X
publication_status: published
publisher: Springer Nature
quality_controlled: '1'
scopus_import: '1'
status: public
title: Adversarial robustness via noise injection in smoothed models
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 53
year: '2022'
...
---
_id: '18232'
abstract:
- lang: eng
text: Cross-modal retrieval aims to retrieve related items across different modalities,
for example, using an image query to retrieve related text. The existing deep
methods ignore both the intra-modal and inter-modal intra-class low-rank structures
when fusing various modalities, which decreases the retrieval performance. In
this paper, two deep models (denoted as ILCMR and Semi-ILCMR) based on intra-class
low-rank regularization are proposed for supervised and semi-supervised cross-modal
retrieval, respectively. Specifically, ILCMR integrates the image network and
text network into a unified framework to learn a common feature space by imposing
three regularization terms to fuse the cross-modal data. First, to align them
in the label space, we utilize semantic consistency regularization to convert
the data representations to probability distributions over the classes. Second,
we introduce an intra-modal low-rank regularization, which encourages the intra-class
samples that originate from the same space to be more relevant in the common feature
space. Third, an inter-modal low-rank regularization is applied to reduce the
cross-modal discrepancy. To enable the low-rank regularization to be optimized
using automatic gradients during network back-propagation, we propose the rank-r
approximation and specify the explicit gradients for theoretical completeness.
In addition to the three regularization terms that rely on label information incorporated
by ILCMR, we propose Semi-ILCMR in the semi-supervised regime, which introduces
a low-rank constraint before projecting the general representations into the common
feature space. Extensive experiments on four public cross-modal datasets demonstrate
the superiority of ILCMR and Semi-ILCMR over other state-of-the-art methods.
article_processing_charge: No
article_type: original
author:
- first_name: Peipei
full_name: Kang, Peipei
last_name: Kang
- first_name: Zehang
full_name: Lin, Zehang
last_name: Lin
- first_name: Zhenguo
full_name: Yang, Zhenguo
last_name: Yang
- first_name: Xiaozhao
full_name: Fang, Xiaozhao
last_name: Fang
- first_name: Alexander
full_name: Bronstein, Alexander
id: 58f3726e-7cba-11ef-ad8b-e6e8cb3904e6
last_name: Bronstein
orcid: 0000-0001-9699-8730
- first_name: Qing
full_name: Li, Qing
last_name: Li
- first_name: Wenyin
full_name: Liu, Wenyin
last_name: Liu
citation:
ama: Kang P, Lin Z, Yang Z, et al. Intra-class low-rank regularization for supervised
and semi-supervised cross-modal retrieval. Applied Intelligence. 2022;52:33-54.
doi:10.1007/s10489-021-02308-3
apa: Kang, P., Lin, Z., Yang, Z., Fang, X., Bronstein, A. M., Li, Q., & Liu,
W. (2022). Intra-class low-rank regularization for supervised and semi-supervised
cross-modal retrieval. Applied Intelligence. Springer Nature. https://doi.org/10.1007/s10489-021-02308-3
chicago: Kang, Peipei, Zehang Lin, Zhenguo Yang, Xiaozhao Fang, Alex M. Bronstein,
Qing Li, and Wenyin Liu. “Intra-Class Low-Rank Regularization for Supervised and
Semi-Supervised Cross-Modal Retrieval.” Applied Intelligence. Springer
Nature, 2022. https://doi.org/10.1007/s10489-021-02308-3.
ieee: P. Kang et al., “Intra-class low-rank regularization for supervised
and semi-supervised cross-modal retrieval,” Applied Intelligence, vol.
52. Springer Nature, pp. 33–54, 2022.
ista: Kang P, Lin Z, Yang Z, Fang X, Bronstein AM, Li Q, Liu W. 2022. Intra-class
low-rank regularization for supervised and semi-supervised cross-modal retrieval.
Applied Intelligence. 52, 33–54.
mla: Kang, Peipei, et al. “Intra-Class Low-Rank Regularization for Supervised and
Semi-Supervised Cross-Modal Retrieval.” Applied Intelligence, vol. 52,
Springer Nature, 2022, pp. 33–54, doi:10.1007/s10489-021-02308-3.
short: P. Kang, Z. Lin, Z. Yang, X. Fang, A.M. Bronstein, Q. Li, W. Liu, Applied
Intelligence 52 (2022) 33–54.
date_created: 2024-10-08T12:56:38Z
date_published: 2022-01-01T00:00:00Z
date_updated: 2024-10-15T07:30:00Z
day: '01'
doi: 10.1007/s10489-021-02308-3
extern: '1'
fulldoi: https://doi.org/10.1007/s10489-021-02308-3
intvolume: ' 52'
language:
- iso: eng
month: '01'
oa_version: None
page: 33-54
publication: Applied Intelligence
publication_identifier:
eissn:
- 1573-7497
issn:
- 0924-669X
publication_status: published
publisher: Springer Nature
quality_controlled: '1'
scopus_import: '1'
status: public
title: Intra-class low-rank regularization for supervised and semi-supervised cross-modal
retrieval
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 52
year: '2022'
...