Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
102 tokens/sec
GPT-4o
59 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
50 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Towards Lifecycle Unlearning Commitment Management: Measuring Sample-level Approximate Unlearning Completeness (2403.12830v2)

Published 19 Mar 2024 in cs.LG and cs.CR

Abstract: By adopting a more flexible definition of unlearning and adjusting the model distribution to simulate training without the targeted data, approximate machine unlearning provides a less resource-demanding alternative to the more laborious exact unlearning methods. Yet, the unlearning completeness of target samples-even when the approximate algorithms are executed faithfully without external threats-remains largely unexamined, raising questions about those approximate algorithms' ability to fulfill their commitment of unlearning during the lifecycle. In this paper, we introduce the task of Lifecycle Unlearning Commitment Management (LUCM) for approximate unlearning and outline its primary challenges. We propose an efficient metric designed to assess the sample-level unlearning completeness. Our empirical results demonstrate its superiority over membership inference techniques in two key areas: the strong correlation of its measurements with unlearning completeness across various unlearning tasks, and its computational efficiency, making it suitable for real-time applications. Additionally, we show that this metric is able to serve as a tool for monitoring unlearning anomalies throughout the unlearning lifecycle, including both under-unlearning and over-unlearning. We apply this metric to evaluate the unlearning commitments of current approximate algorithms. Our analysis, conducted across multiple unlearning benchmarks, reveals that these algorithms inconsistently fulfill their unlearning commitments due to two main issues: 1) unlearning new data can significantly affect the unlearning utility of previously requested data, and 2) approximate algorithms fail to ensure equitable unlearning utility across different groups. These insights emphasize the crucial importance of LUCM throughout the unlearning lifecycle. We will soon open-source our newly developed benchmark.

PDF HTML Abstract
Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Bookmark Chat Bubble Oval Streamline Icon: https://streamlinehq.com Chat (Pro)
Definition Search Book Streamline Icon: https://streamlinehq.com
References (34)
  1. Influence functions in deep learning are fragile. In 9th International Conference on Learning Representations, ICLR 2021, Virtual Event, Austria, May 3-7, 2021. OpenReview.net, 2021.
  2. Evaluating machine unlearning via epistemic uncertainty. CoRR, abs/2208.10836, 2022.
  3. Five years of the right to be forgotten. In Lorenzo Cavallaro, Johannes Kinder, XiaoFeng Wang, and Jonathan Katz, editors, Proceedings of the 2019 ACM SIGSAC Conference on Computer and Communications Security, CCS 2019, London, UK, November 11-15, 2019, pages 959–972. ACM, 2019.
  4. Machine unlearning. In 42nd IEEE Symposium on Security and Privacy, SP 2021, San Francisco, CA, USA, 24-27 May 2021, pages 141–159. IEEE, 2021.
  5. What does it mean for a language model to preserve privacy? In FAccT ’22: 2022 ACM Conference on Fairness, Accountability, and Transparency, Seoul, Republic of Korea, June 21 - 24, 2022, pages 2280–2292. ACM, 2022.
  6. "you might also like: " privacy risks of collaborative filtering. In 32nd IEEE Symposium on Security and Privacy, SP 2011, 22-25 May 2011, Berkeley, California, USA, pages 231–246. IEEE Computer Society, 2011.
  7. Towards making systems forget with machine unlearning. In 2015 IEEE Symposium on Security and Privacy, SP 2015, San Jose, CA, USA, May 17-21, 2015, pages 463–480. IEEE Computer Society, 2015.
  8. Membership inference attacks from first principles. In 43rd IEEE Symposium on Security and Privacy, SP 2022, San Francisco, CA, USA, May 22-26, 2022, pages 1897–1914. IEEE, 2022.
  9. Quantifying memorization across neural language models. In The Eleventh International Conference on Learning Representations, ICLR 2023, Kigali, Rwanda, May 1-5, 2023. OpenReview.net, 2023.
  10. The privacy onion effect: Memorization is relative. In Sanmi Koyejo, S. Mohamed, A. Agarwal, Danielle Belgrave, K. Cho, and A. Oh, editors, Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, NeurIPS 2022, New Orleans, LA, USA, November 28 - December 9, 2022, 2022.
  11. Boundary unlearning: Rapid forgetting of deep networks via shifting the decision boundary. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023, Vancouver, BC, Canada, June 17-24, 2023, pages 7766–7775. IEEE, 2023.
  12. When machine unlearning jeopardizes privacy. In Yongdae Kim, Jong Kim, Giovanni Vigna, and Elaine Shi, editors, CCS ’21: 2021 ACM SIGSAC Conference on Computer and Communications Security, Virtual Event, Republic of Korea, November 15 - 19, 2021, pages 896–911. ACM, 2021.
  13. Fast machine unlearning without retraining through selective synaptic dampening. CoRR, abs/2308.07707, 2023.
  14. I know where you parked last summer : Automated reverse engineering and privacy analysis of modern cars. In 2020 IEEE Symposium on Security and Privacy, SP 2020, San Francisco, CA, USA, May 18-21, 2020, pages 1401–1415. IEEE, 2020.
  15. Towards adversarial evaluations for inexact machine unlearning. 2023.
  16. Eternal sunshine of the spotless net: Selective forgetting in deep networks. In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2020, Seattle, WA, USA, June 13-19, 2020, pages 9301–9309. Computer Vision Foundation / IEEE, 2020.
  17. Forgetting outside the box: Scrubbing deep networks of information accessible from input-output observations. In Computer Vision - ECCV 2020 - 16th European Conference, Glasgow, UK, August 23-28, 2020, Proceedings, Part XXIX, volume 12374 of Lecture Notes in Computer Science, pages 383–398. Springer, 2020.
  18. Amnesiac machine learning. In Thirty-Fifth AAAI Conference on Artificial Intelligence, AAAI 2021, Thirty-Third Conference on Innovative Applications of Artificial Intelligence, IAAI 2021, The Eleventh Symposium on Educational Advances in Artificial Intelligence, EAAI 2021, Virtual Event, February 2-9, 2021, pages 11516–11524. AAAI Press, 2021.
  19. EMA: auditing data removal from trained models. In Marleen de Bruijne, Philippe C. Cattin, Stéphane Cotin, Nicolas Padoy, Stefanie Speidel, Yefeng Zheng, and Caroline Essert, editors, Medical Image Computing and Computer Assisted Intervention - MICCAI 2021 - 24th International Conference, Strasbourg, France, September 27 - October 1, 2021, Proceedings, Part V, volume 12905 of Lecture Notes in Computer Science, pages 793–803. Springer, 2021.
  20. How to combine membership-inference attacks on multiple updated machine learning models. Proc. Priv. Enhancing Technol., 2023(3):211–232, 2023.
  21. Proof-of-learning: Definitions and practice. In 42nd IEEE Symposium on Security and Privacy, SP 2021, San Francisco, CA, USA, 24-27 May 2021, pages 1039–1056. IEEE, 2021.
  22. Overcoming catastrophic forgetting in neural networks. Proceedings of the National Academy of Sciences, 114(13):3521–3526, 2017.
  23. Detecting outliers: Do not use standard deviation around the mean, use absolute deviation around the median. Journal of Experimental Social Psychology, 49(4):764–766, 2013.
  24. Have you forgotten? A method to assess if machine learning models have forgotten data. In Anne L. Martel, Purang Abolmaesumi, Danail Stoyanov, Diana Mateus, Maria A. Zuluaga, S. Kevin Zhou, Daniel Racoceanu, and Leo Joskowicz, editors, Medical Image Computing and Computer Assisted Intervention - MICCAI 2020 - 23rd International Conference, Lima, Peru, October 4-8, 2020, Proceedings, Part I, volume 12261 of Lecture Notes in Computer Science, pages 95–105. Springer, 2020.
  25. Learn to forget: Machine unlearning via neuron masking. IEEE Trans. Dependable Secur. Comput., 20(4):3194–3207, 2023.
  26. Deep unlearning via randomized conditionally independent hessians. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022, New Orleans, LA, USA, June 18-24, 2022, pages 10412–10421. IEEE, 2022.
  27. Which privacy and security attributes most impact consumers’ risk perception and willingness to purchase iot devices? In 42nd IEEE Symposium on Security and Privacy, SP 2021, San Francisco, CA, USA, 24-27 May 2021, pages 519–536. IEEE, 2021.
  28. Stuart L Pardau. The california consumer privacy act: towards a european-style privacy regime in the united states. J. Tech. L. & Pol’y, 23:68, 2018.
  29. Remember what you want to forget: Algorithms for machine unlearning. In Marc’Aurelio Ranzato, Alina Beygelzimer, Yann N. Dauphin, Percy Liang, and Jennifer Wortman Vaughan, editors, Advances in Neural Information Processing Systems 34: Annual Conference on Neural Information Processing Systems 2021, NeurIPS 2021, December 6-14, 2021, virtual, pages 18075–18086, 2021.
  30. Membership inference attacks against machine learning models. In 2017 IEEE Symposium on Security and Privacy, SP 2017, San Jose, CA, USA, May 22-26, 2017, pages 3–18. IEEE Computer Society, 2017.
  31. Unrolling SGD: understanding factors influencing machine unlearning. In 7th IEEE European Symposium on Security and Privacy, EuroS&P 2022, Genoa, Italy, June 6-10, 2022, pages 303–319. IEEE, 2022.
  32. On the necessity of auditable algorithmic definitions for machine unlearning. In 31st USENIX Security Symposium (USENIX Security 22), pages 4007–4022, 2022.
  33. Machine unlearning of features and labels. 2023.
  34. Deltagrad: Rapid retraining of machine learning models. In Proceedings of the 37th International Conference on Machine Learning, ICML 2020, 13-18 July 2020, Virtual Event, volume 119 of Proceedings of Machine Learning Research, pages 10355–10366. PMLR, 2020.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (5)
  1. Cheng-Long Wang (10 papers)
  2. Qi Li (354 papers)
  3. Zihang Xiang (10 papers)
  4. Di Wang (407 papers)
  5. Yinzhi Cao (26 papers)
Citations (1)
View on Semantic Scholar
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets