Katherine Lee
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Katherine Lee has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
Chiyuan Zhang
2 shared publications- Daphne Ippolito
University of Pennsylvania
2 shared publications - Florian Tramèr
ETH Zurich, Zurich, Switzerland
2 shared publications - Matthew Jagielski
Anthropic, Boston, MA, USA
2 shared publications Nicholas Carlini
2 shared publications- A. Feder Cooper
The GenLaw Center, San Francisco, CA, USA, Cornell University, Ithaca, NY, USA
1 shared publication
Latest Publications
- Open Access
- 12 March 2024
- Published by Association for Computing Machinery (ACM)
- p. 48-63
Talkin' 'Bout AI Generation
- 12 March 2024
- Published by Association for Computing Machinery (ACM)
- p. 48-63
Open AccessTalkin' 'Bout AI Generation
- Open Access
Measuring Forgetting of Memorized Training Examples
Machine learning models exhibit two seemingly contradictory phenomena: training data memorization, and various forms of forgetting. In memorization, models overfit specific training examples and become susceptible to privacy attacks. In forgetting, examples which appeared early in training are forgotten by the end. In this work, we connect these phenomena. We propose a technique to measure to what extent models "forget" the specifics of training examples, becoming less susceptible to privacy attacks on examples they have not seen recently. We show that, while non-convex models can memorize data forever in the worst-case, standard image, speech, and language models empirically do forget examples over time. We identify nondeterminism as a potential explanation, showing that deterministically trained models do not forget. Our results suggest that examples seen early when training with extremely large datasets - for instance those examples used to pre-train a model - may observe privacy benefits at the expense of examples seen later.
Open AccessMeasuring Forgetting of Memorized Training Examples
Machine learning models exhibit two seemingly contradictory phenomena: training data memorization, and various forms of forgetting. In memorization, models overfit specific training examples and become susceptible to privacy attacks. In forgetting, examples which appeared early in training are forgotten by the end. In this work, we connect these phenomena. We propose a technique to measure to what extent models "forget" the specifics of training examples, becoming less susceptible to privacy attacks on examples they have not seen recently. We show that, while non-convex models can memorize data forever in the worst-case, standard image, speech, and language models empirically do forget examples over time. We identify nondeterminism as a potential explanation, showing that deterministically trained models do not forget. Our results suggest that examples seen early when training with extremely large datasets - for instance those examples used to pre-train a model - may observe privacy benefits at the expense of examples seen later.
- Top Cited PapersOpen Access
- 1 January 2023
- Published by Association for Computational Linguistics (ACL)
- p. 28-53
Preventing Generation of Verbatim Memorization in Language Models Gives a False Sense of Privacy
Daphne Ippolito, Florian Tramer, Milad Nasr, Chiyuan Zhang, Matthew Jagielski, Katherine Lee, Christopher Choquette Choo, Nicholas Carlini. Proceedings of the 16th International Natural Language Generation Conference. 2023.
- 1 January 2023
- Published by Association for Computational Linguistics (ACL)
- p. 28-53
Top Cited PapersOpen AccessPreventing Generation of Verbatim Memorization in Language Models Gives a False Sense of Privacy
Daphne Ippolito, Florian Tramer, Milad Nasr, Chiyuan Zhang, Matthew Jagielski, Katherine Lee, Christopher Choquette Choo, Nicholas Carlini. Proceedings of the 16th International Natural Language Generation Conference. 2023.
- 1 February 2020
- Published by Elsevier in Journal of Pain and Symptom Management
- Vol. 59 (2) , 537-538
An Artificial Intelligence Algorithm to Identify Documented Symptoms in Patients with Heart Failure who Received Cardiac Resynchronization Therapy (S717)
- 1 February 2020
- Published by Elsevier in Journal of Pain and Symptom Management
- Vol. 59 (2) , 537-538
An Artificial Intelligence Algorithm to Identify Documented Symptoms in Patients with Heart Failure who Received Cardiac Resynchronization Therapy (S717)