The Library
Towards unbounded machine unlearning
Tools
Kurmanji, M., Triantafillou, Peter, Hayes, Jamie and Triantafillou, E. (2023) Towards unbounded machine unlearning. In: International Conference on Neural Information Processing Systems, NeurIPS, New Orleans, 10-16 Dec 2023 (In Press)
An open access version can be found in:
Official URL: https://neurips.cc/virtual/2023/poster/71789
Abstract
Deep machine unlearning is the problem of 'removing' from a trained neural network a subset of its training set. This problem is very timely and has many applications, including the key tasks of removing biases (RB), resolving confusion (RC) (caused by mislabelled data in trained models), as well as allowing users to exercise their 'right to be forgotten' to protect User Privacy (UP). This paper is the first, to our knowledge, to study unlearning for different applications (RB, RC, UP), with the view that each has its own desiderata, definitions for 'forgetting' and associated metrics for forget quality. We also propose SCRUB, a novel unlearning algorithm, which is the only method that is consistently a top performer for forget quality across the different application-dependent metrics for RB, RC, and UP. At the same time, SCRUB is also consistently a top performer on metrics that measure model utility (i.e. accuracy on retained data and generalization), and efficiency/scalability. The above are substantiated through a comprehensive empirical evaluation against previous state-of-the-art on different datasets and architectures.
Item Type: | Conference Item (Poster) | ||||||
---|---|---|---|---|---|---|---|
Divisions: | Faculty of Science, Engineering and Medicine > Science > Computer Science | ||||||
Official Date: | December 2023 | ||||||
Dates: |
|
||||||
Status: | Peer Reviewed | ||||||
Publication Status: | In Press | ||||||
Access rights to Published version: | Restricted or Subscription Access | ||||||
Conference Paper Type: | Poster | ||||||
Title of Event: | International Conference on Neural Information Processing Systems, NeurIPS | ||||||
Type of Event: | Conference | ||||||
Location of Event: | New Orleans | ||||||
Date(s) of Event: | 10-16 Dec 2023 | ||||||
Related URLs: | |||||||
Open Access Version: |
Request changes or add full text files to a record
Repository staff actions (login required)
View Item |