The DOI matches its Crossref record; 17 of the 17 references it cites were found in a registry or an index.
Model-Reuse Attacks on Deep Learning Systems
Open accessAbstract
Many of today's machine learning (ML) systems are built by reusing an array of, often pre-trained, primitive models, each fulfilling distinct functionality (e.g., feature extraction). The increasing use of primitive models significantly simplifies and expedites the development cycles of ML systems. Yet, because most of such models are contributed and maintained by untrusted sources, their lack of standardization or regulation entails profound security implications, about which little is known thus far. In this paper, we demonstrate that malicious primitive models pose immense threats to the security of ML systems. We present a broad class of model-reuse attacks wherein maliciously crafted models trigger host ML systems to misbehave on targeted inputs in a highly predictable manner. By empirically studying four deep learning systems (including both individual and ensemble systems) used in skin cancer screening, speech recognition, face verification, and autonomous steering, we show that such attacks are (i) effective - the host systems misbehave on the targeted inputs as desired by the adversary with high probability, (ii) evasive - the malicious models function indistinguishably from their benign counterparts on non-targeted inputs, (iii) elastic - the malicious models remain effective regardless of various system design choices and tuning strategies, and (iv) easy - the adversary needs little prior knowledge about the data used for system tuning or inference. We provide analytical justification for the effectiveness of model-reuse attacks, which points to the unprecedented complexity of today's primitive models. This issue thus seems fundamental to many ML systems. We further discuss potential countermeasures and their challenges, which lead to several promising research directions.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Self-citationImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Dermatologist-level classification of skin cancer with deep neural networks
Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017) · Nature · doi:10.1038/nature21056 · Cited by 14,362
Explaining and Harnessing Adversarial Examples
Goodfellow, I. J., Shlens, J., & Szegedy, C. (2014) · arXiv · doi:10.48550/arxiv.1412.6572 · Cited by 10,381
Towards Evaluating the Robustness of Neural Networks
Carlini, N., & Wagner, D. (2017) · 2017 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2017.49 · Cited by 8,859
The Limitations of Deep Learning in Adversarial Settings
Papernot, N., McDaniel, P., Jha, S., Fredrikson, M., Celik, Z. B., & Swami, A. (2016) · 2016 IEEE European Symposium on Security and Privacy (EuroS&P) · doi:10.1109/eurosp.2016.36 · Cited by 3,993
VGGFace2: A Dataset for Recognising Faces across Pose and Age
Cao, Q., Shen, L., Xie, W., Parkhi, O. M., & Zisserman, A. (2018) · 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018) · doi:10.1109/fg.2018.00020 · Cited by 3,032
Understanding deep learning requires rethinking generalization
Zhang, C., Bengio, S., Hardt, M., Recht, B., & Vinyals, O. (2016) · arXiv · doi:10.48550/arxiv.1611.03530 · Cited by 2,838 · cited 1×
“Recent studies [64] have further provided both empirical and theoretical evidence that the effective capacities of many DNNs are sufficient for “memorizing” entire training sets.”
BackgroundTrojaning Attack on Neural Networks
Liu, Y., Ma, S., Aafer, Y., Lee, W.-C., Zhai, J., Wang, W., & Zhang, X. (2018) · Proceedings 2018 Network and Distributed System Security Symposium · doi:10.14722/ndss.2018.23291 · Cited by 1,270
Self-citation
Cited by
How to backdoor federated learning
Bagdasaryan, E., Veit, A., Hua, Y., Estrin, D., & Shmatikov, V. (2018) · arXiv · doi:10.48550/arxiv.1807.00459 · Cited by 1,495
BadNets: Evaluating Backdooring Attacks on Deep Neural Networks
Gu, T., Liu, K., Dolan-Gavitt, B., & Garg, S. (2019) · IEEE Access · doi:10.1109/access.2019.2909068 · Cited by 1,238
“A related body of work has looked causative attacks on deep learning [39]–[41], but assumes very different attack goals compared to backdooring.”
BackgroundSTRIP: a defence against trojan attacks on deep neural networks
Gao, Y., Xu, C., Wang, D., Chen, S., Ranasinghe, D. C., & Nepal, S. (2019) · Proceedings of the 35th Annual Computer Security Applications Conference · doi:10.1145/3359789.3359790 · Cited by 707
ABS: Scanning Neural Networks for Back-doors by Artificial Brain Stimulation
Liu, Y., Lee, W.-C., Tao, G., Ma, S., Aafer, Y., & Zhang, X. (2019) · Proceedings of the 2019 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3319535.3363216 · Cited by 404
Self-citationWeight Poisoning Attacks on Pretrained Models
Kurita, K., Michel, P., & Neubig, G. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.249 · Cited by 354
Software Engineering for AI-Based Systems: A Survey
Martínez-Fernández, S., Bogner, J., Franch, X., Oriol, M., Siebert, J., Trendowicz, A., Vollmer, A. M., & Wagner, S. (2022) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/3487043 · Cited by 335
Dataset Security for Machine Learning: Data Poisoning, Backdoor Attacks, and Defenses
Goldblum, M., Tsipras, D., Xie, C., Chen, X., Schwarzschild, A., Song, D., Madry, A., Li, B., & Goldstein, T. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3162397 · Cited by 362
“An attacker can break many transfer learned systems by collecting a large battery of standard models and creating poisons that cause feature collisions for all of them [38].”
BackgroundSentiNet: Detecting Localized Universal Attacks Against Deep Learning Systems
Chou, E., Tramer, F., & Pellegrino, G. (2020) · 2020 IEEE Security and Privacy Workshops (SPW) · doi:10.1109/spw50608.2020.00025 · Cited by 281
Lwakatare, L. E., Raj, A., Crnkovic, I., Bosch, J., & Olsson, H. H. (2020) · Information and Software Technology · doi:10.1016/j.infsof.2020.106368 · Cited by 241
When Machine Unlearning Jeopardizes Privacy
Chen, M., Zhang, Z., Wang, T., Backes, M., Humbert, M., & Zhang, Y. (2021) · Proceedings of the 2021 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3460120.3484756 · Cited by 195
Self-citationPrivacy Risks of General-Purpose Language Models
Pan, X., Zhang, M., Ji, S., & Yang, M. (2020) · 2020 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp40000.2020.00095 · Cited by 188
“This observation is also consistent with the conclusion in [34]: complicated models tend to enlarge the attack surface.”
ResultSelf-citationMARNet: Backdoor Attacks Against Cooperative Multi-Agent Reinforcement Learning
Chen, Y., Zheng, Z., & Gong, X. (2023) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2022.3207429 · Cited by 32
“Existing defense methods for backdoor attacks are mainly designed for deep neural networks [30], [31], [32], [33], and there is a lack of defense for backdoor attacks against DRL.”
Semi-Targeted Model Poisoning Attack on Federated Learning via Backward Error Analysis
Sun, Y., Ochiai, H., & Sakuma, J. (2022) · 2022 International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn55064.2022.9891990 · Cited by 28
A Synergetic Attack against Neural Network Classifiers combining Backdoor and Adversarial Examples
Liu, G., Khalil, I., Khreishah, A., & Phan, N. (2021) · 2021 IEEE International Conference on Big Data (Big Data) · doi:10.1109/bigdata52589.2021.9671964 · Cited by 11
“This setting has been shown to be realistic [20], since: (1) Model re-usability is important in many applications to reduce the tremendous amount of time and computational resources for model training.”
Background
Metadata
- Authors
- Ji, Y., Zhang, X., Ji, S., Luo, X., & Wang, T.
- Venue
- Proceedings of the 2018 ACM SIGSAC Conference on Computer and Communications Security
- Publisher
- ACM
- Published
- 2018-10-15
- Pages
- 349-363
- OpenAlex
- W2897865027
- Cited by
- 167
- References
- 17
- FWCI
- 8.53
- Type
- conference paper
- Language
- en
- Open access
- hybrid
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.