Neurips 2023-machine unlearning, 2023

[PDF] neurips.cc

Towards unbounded machine unlearning

M Kurmanji, P Triantafillou, J Hayes… - Advances in neural …, 2024 - proceedings.neurips.cc
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 …
被引用次数：147 相关文章 所有 7 个版本
[PDF] arxiv.org

Tofu: A task of fictitious unlearning for llms

P Maini, Z Feng, A Schwarzschild, ZC Lipton… - arXiv preprint arXiv …, 2024 - arxiv.org
Large language models trained on massive corpora of data from the web can memorize and
reproduce sensitive or private data raising both legal and ethical concerns. Unlearning, or …
被引用次数：24 相关文章 所有 5 个版本
[PDF] arxiv.org

Inexact unlearning needs more careful evaluations to avoid a false sense of privacy

J Hayes, I Shumailov, E Triantafillou, A Khalifa… - arXiv preprint arXiv …, 2024 - arxiv.org
The high cost of model training makes it increasingly desirable to develop techniques for
unlearning. These techniques seek to remove the influence of a training example without …
被引用次数：24 相关文章 所有 2 个版本
[PDF] arxiv.org

Unlearning in-vs. out-of-distribution data in LLMs under gradient-based method

T Baluta, P Lamblin, D Tarlow, F Pedregosa… - arXiv preprint arXiv …, 2024 - arxiv.org
Machine unlearning aims to solve the problem of removing the influence of selected training
examples from a learned model. Despite the increasing attention to this problem, it remains …
相关文章 所有 3 个版本
[PDF] arxiv.org

Data Selection for Transfer Unlearning

NM Sepahvand, V Dumoulin, E Triantafillou… - arXiv preprint arXiv …, 2024 - arxiv.org
As deep learning models are becoming larger and data-hungrier, there are growing ethical,
legal and technical concerns over use of data: in practice, agreements on data use may …
被引用次数：3 相关文章 所有 2 个版本
[PDF] arxiv.org

Deep Unlearn: Benchmarking Machine Unlearning

XF Cadet, A Borovykh, M Malekzadeh… - arXiv preprint arXiv …, 2024 - arxiv.org
Machine unlearning (MU) aims to remove the influence of particular data points from the
learnable parameters of a trained machine learning model. This is a crucial capability in light …
相关文章 所有 3 个版本