Dataset Security for Machine Learning: Data Poisoning, Backdoor Attacks, and Defenses
IEEE Transactions on Pattern Analysis and Machine Intelligence · 01.02.2023 · 45 (2) · 1563-1580
DOI: 10.1109/tpami.2022.3162397 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, erschienen am 01.02.2023. Verfasst von Micah Goldblum und 8 weiteren Autor:innen. Die Arbeit wurde bisher 299-mal zitiert (Crossref) und nennt 205 Literaturangaben. Gefördert von: Defense Advanced Research Projects Agency, QED4RML, D3M programs.
Abstract (Originalsprache)
As machine learning systems grow in scale, so do their training data requirements, forcing practitioners to automate and outsource the curation of training data in order to achieve state-of-the-art performance. The absence of trustworthy human supervision over the data collection process exposes organizations to security vulnerabilities; training data can be manipulated to control and degrade the downstream behaviors of learned models. The goal of this work is to systematically categorize and discuss a wide range of dataset vulnerabilities and exploits, approaches for defending against these threats, and an array of open problems in this space.
Förderung
- Defense Advanced Research Projects Agency
- QED4RML
- D3M programs
- National Science Foundation
- Amazon Research Award Program
- Berkeley DeepDrive
- Facebook PhD Fellowship
- ONR MURI Program
Zitieren
APA
Goldblum, M., Tsipras, D., Xie, C., Chen, X., Schwarzschild, A., Song, D., Madry, A., Li, B., & Goldstein, T. (2023). Dataset Security for Machine Learning: Data Poisoning, Backdoor Attacks, and Defenses. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(2), 1563-1580. https://doi.org/10.1109/tpami.2022.3162397
Vancouver
Goldblum M, Tsipras D, Xie C, Chen X, Schwarzschild A, Song D, et al. Dataset Security for Machine Learning: Data Poisoning, Backdoor Attacks, and Defenses. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2023;45(2):1563-1580. doi:10.1109/tpami.2022.3162397
BibTeX
@article{Goldblum2023,
author = {Goldblum, Micah and Tsipras, Dimitris and Xie, Chulin and Chen, Xinyun and Schwarzschild, Avi and Song, Dawn and Madry, Aleksander and Li, Bo and Goldstein, Tom},
title = {Dataset Security for Machine Learning: Data Poisoning, Backdoor Attacks, and Defenses},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {2023},
volume = {45},
number = {2},
pages = {1563--1580},
doi = {10.1109/tpami.2022.3162397}
}
EUVIMED