The DOI matches its Crossref record; 39 of the 39 references it cites were found in a registry or an index.
Unrolling SGD: Understanding Factors Influencing Machine Unlearning
Open accessAbstract
Machine unlearning is the process through which a deployed machine learning model is made to forget about some of its training data points. While naively retraining the model from scratch is an option, it is almost always associated with large computational overheads for deep learning models. Thus, several approaches to approximately unlearn have been proposed along with corresponding metrics that formalize what it means for a model to forget about a data point. In this work, we first taxonomize approaches and metrics of approximate unlearning. As a result, we identify verification error, i.e., the$\ell_{2}$difference between the weights of an approximately unlearned and a naively retrained model, as an approximate unlearning metric that should be optimized for as it subsumes a large class of other metrics. We theoretically analyze the canonical training algorithm, stochastic gradient descent (SGD), to surface the variables which are relevant to reducing the verification error of approximate unlearning for SGD. From this analysis, we first derive an easy-to-compute proxy for verification error (termed unlearning error). The analysis also informs the design of a new training objective penalty that limits the overall change in weights during SGD and as a result facilitates approximate unlearning with lower verification error. We validate our theoretical work through an empirical evaluation on learning with CIFAR-10, CIFAR-100, and IMDB sentiment analysis.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148
ImageNet 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
Learning Multiple Layers of Features from Tiny Images
Krizhevsky, A. (2024) · doi:10.57702/zp44cu3g · Cited by 23,889
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Deep Learning with Differential Privacy
Abadi, M., Chu, A., Goodfellow, I., McMahan, H. B., Mironov, I., Talwar, K., & Zhang, L. (2016) · Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/2976749.2978318 · Cited by 6,578
The Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2014) · Foundations and Trends® in Theoretical Computer Science · doi:10.1561/0400000042 · Cited by 5,557
Membership Inference Attacks Against Machine Learning Models
Shokri, R., Stronati, M., Song, C., & Shmatikov, V. (2017) · 2017 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2017.41 · Cited by 4,754
DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter
Sanh, V., Debut, L., Chaumond, J., & Wolf, T. (2019) · arXiv · doi:10.48550/arxiv.1910.01108 · Cited by 4,653
The Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2013) · doi:10.1561/9781601988195 · Cited by 3,856
Research works printed but not linked: 1
- ba, 2021, Csc421/2516 lecture 3: Automatic differentiation & distributed representations, Lecture Notes WS
Cited by
Sorted by citations
A Survey of Machine Unlearning
Nguyên, T. T., Huynh, T. T., Ren, Z. X., Nguyen, P. L., Liew, A. W., Yin, H., & Nguyen, Q. V. H. (2025) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3749987 · Cited by 540
Xu, H., Zhu, T., Zhang, L., Zhou, W., & Yu, P. S. (2024) · ACM Computing Surveys · doi:10.1145/3603620 · Cited by 164
“[37] discussed the semantics behind unlearning and reviewed existing unlearning schemes based on logits, weights, and weight distributions.”
BackgroundMachine Unlearning: Solutions and Challenges
Xu, J., Wu, Z., Wang, C., & Jia, X. (2024) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2024.3379240 · Cited by 139
Rethinking machine unlearning for large language models
Liu, S., Yao, Y., Jia, J., Casper, S., Baracaldo, N., Hase, P., Yao, Y., Liu, C. Y., Xu, X., Li, H., et al. (2025) · Nature Machine Intelligence · doi:10.1038/s42256-025-00985-0 · Cited by 374
A Review on Machine Unlearning
Zhang, H., Nakamura, T., Isohara, T., & Sakurai, K. (2023) · SN Computer Science · doi:10.1007/s42979-023-01767-4 · Cited by 81
Metadata
- Authors
- Thudi, A., Deza, G., Chandrasekaran, V., & Papernot, N.
- Affiliations
- University of Toronto (CA)
- University of Wisconsin–Madison (US)
- Venue
- 2022 IEEE 7th European Symposium on Security and Privacy (EuroS&P)
- Publisher
- IEEE
- Published
- 2022-06
- Pages
- 303-319
- OpenAlex
- W3202838631
- Cited by
- 135
- References
- 38
- Citations by year
- 2026: 45 · 2025: 54 · 2024: 22 · 2023: 12 · 2022: 1
- FWCI
- 12.30
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE European Symposium on Security and Privacy (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- DARPA
- NSERC
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Artificial Intelligence › Stochastic Gradient Optimization Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Keywords
- machine unlearning, approximate unlearning, stochastic gradient descent, weight change, deep learning
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
