Redirection for Erasing Memory (REM): Towards a universal unlearning method for corrupted data
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915770754138112 |
|---|---|
| author | Schoepf, Stefan Mozer, Michael Curtis Mitchell, Nicole Elyse Brintrup, Alexandra Kaissis, Georgios Kairouz, Peter Triantafillou, Eleni |
| author_facet | Schoepf, Stefan Mozer, Michael Curtis Mitchell, Nicole Elyse Brintrup, Alexandra Kaissis, Georgios Kairouz, Peter Triantafillou, Eleni |
| contents | Machine unlearning is studied for a multitude of tasks, but specialization of unlearning methods to particular tasks has made their systematic comparison challenging. To address this issue, we propose a conceptual space to characterize diverse corrupted data unlearning tasks in vision classifiers. This space is described by two dimensions, the discovery rate (the fraction of the corrupted data that are known at unlearning time) and the statistical regularity of the corrupted data (from random exemplars to shared concepts). Methods proposed previously have been targeted at portions of this space and-we show-fail predictably outside these regions. We propose a novel method, Redirection for Erasing Memory (REM), whose key feature is that corrupted data are redirected to dedicated neurons introduced at unlearning time and then discarded or deactivated to suppress the influence of corrupted data. REM performs strongly across the space of tasks, in contrast to prior SOTA methods that fail outside the regions for which they were designed. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_17730 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Redirection for Erasing Memory (REM): Towards a universal unlearning method for corrupted data Schoepf, Stefan Mozer, Michael Curtis Mitchell, Nicole Elyse Brintrup, Alexandra Kaissis, Georgios Kairouz, Peter Triantafillou, Eleni Machine Learning Machine unlearning is studied for a multitude of tasks, but specialization of unlearning methods to particular tasks has made their systematic comparison challenging. To address this issue, we propose a conceptual space to characterize diverse corrupted data unlearning tasks in vision classifiers. This space is described by two dimensions, the discovery rate (the fraction of the corrupted data that are known at unlearning time) and the statistical regularity of the corrupted data (from random exemplars to shared concepts). Methods proposed previously have been targeted at portions of this space and-we show-fail predictably outside these regions. We propose a novel method, Redirection for Erasing Memory (REM), whose key feature is that corrupted data are redirected to dedicated neurons introduced at unlearning time and then discarded or deactivated to suppress the influence of corrupted data. REM performs strongly across the space of tasks, in contrast to prior SOTA methods that fail outside the regions for which they were designed. |
| title | Redirection for Erasing Memory (REM): Towards a universal unlearning method for corrupted data |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2505.17730 |