Improving Unlearning with Model Updates Probably Aligned with Gradients

• URL: http://arxiv.org/abs/2511.02435v1
• Date: Tue, 04 Nov 2025 10:07:35 GMT
• Title: Improving Unlearning with Model Updates Probably Aligned with Gradients
• Authors: Virgile Dine, Teddy Furon, Charly Faure,
• Abstract summary: We introduce the concept of feasible updates as the model's parameter update directions.<n>Our design of feasible updates is based on masking, ie a careful selection of the model's parameters worth updating.<n>The technique can be plugged in, as an add-on, to any first-order approximate unlearning methods.
• Score: 18.695608729905544
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: We formulate the machine unlearning problem as a general constrained optimization problem. It unifies the first-order methods from the approximate machine unlearning literature. This paper then introduces the concept of feasible updates as the model's parameter update directions that help with unlearning while not degrading the utility of the initial model. Our design of feasible updates is based on masking, \ie\ a careful selection of the model's parameters worth updating. It also takes into account the estimation noise of the gradients when processing each batch of data to offer a statistical guarantee to derive locally feasible updates. The technique can be plugged in, as an add-on, to any first-order approximate unlearning methods. Experiments with computer vision classifiers validate this approach.

Related papers

• Can Gradient Descent Simulate Prompting? [56.60154660021178]
  gradient updates the effects of conditioning on new information.<n> gradient descent training recovers some (and occasionally all) of prompted model performance.<n>Results suggest new avenues for long-context modeling.
  arXiv  Detail & Related papers  (2025-06-26T04:06:20Z)
• Sharpness-Aware Parameter Selection for Machine Unlearning [6.397490580631141]
  It often happens that some sensitive personal information, such as credit card numbers or passwords, are mistakenly incorporated in the training of machine learning models and need to be removed afterwards.<n>There have been various machine unlearning techniques proposed in the literature to address this problem.<n>Most of the proposed methods revolve around removing individual data samples from a trained model.<n>While the existing methods for these tasks do the unlearning task by updating the whole set of model parameters or only the last layer of the model, we show that there are a subset of model parameters that have the largest contribution in the unlearning target features.
  arXiv  Detail & Related papers  (2025-04-08T19:41:07Z)
• Bayesian Test-Time Adaptation for Vision-Language Models [51.93247610195295]
  Test-time adaptation with pre-trained vision-language models, such as CLIP, aims to adapt the model to new, potentially out-of-distribution test data.<n>We propose a novel approach, textbfBayesian textbfClass textbfAdaptation (BCA), which in addition to continuously updating class embeddings to adapt likelihood, also uses the posterior of incoming samples to continuously update the prior for each class embedding.
  arXiv  Detail & Related papers  (2025-03-12T10:42:11Z)
• LoRA Unlearns More and Retains More (Student Abstract) [0.0]
  PruneLoRA reduces the need for large-scale parameter updates by applying low-rank updates to the model. We leverage LoRA to selectively modify a subset of the pruned model's parameters, thereby reducing the computational cost, memory requirements and improving the model's ability to retain performance on the remaining classes.
  arXiv  Detail & Related papers  (2024-11-16T16:47:57Z)
• Attribute-to-Delete: Machine Unlearning via Datamodel Matching [65.13151619119782]
  Machine unlearning -- efficiently removing a small "forget set" training data on a pre-divertrained machine learning model -- has recently attracted interest. Recent research shows that machine unlearning techniques do not hold up in such a challenging setting.
  arXiv  Detail & Related papers  (2024-10-30T17:20:10Z)
• Learn to Unlearn for Deep Neural Networks: Minimizing Unlearning Interference with Gradient Projection [56.292071534857946]
  Recent data-privacy laws have sparked interest in machine unlearning. Challenge is to discard information about the forget'' data without altering knowledge about remaining dataset. We adopt a projected-gradient based learning method, named as Projected-Gradient Unlearning (PGU) We provide empirically evidence to demonstrate that our unlearning method can produce models that behave similar to models retrained from scratch across various metrics even when the training dataset is no longer accessible.
  arXiv  Detail & Related papers  (2023-12-07T07:17:24Z)
• Update Compression for Deep Neural Networks on the Edge [33.57905298104467]
  An increasing number of AI applications involve the execution of deep neural networks (DNNs) on edge devices. Many practical reasons motivate the need to update the DNN model on the edge device post-deployment. We develop a simple approach based on matrix factorisation to compress the model update.
  arXiv  Detail & Related papers  (2022-03-09T04:20:43Z)
• Machine Unlearning of Features and Labels [72.81914952849334]
  We propose first scenarios for unlearning and labels in machine learning models. Our approach builds on the concept of influence functions and realizes unlearning through closed-form updates of model parameters.
  arXiv  Detail & Related papers  (2021-08-26T04:42:24Z)
• Self-Tuning Stochastic Optimization with Curvature-Aware Gradient Filtering [53.523517926927894]
  We explore the use of exact per-sample Hessian-vector products and gradients to construct self-tuning quadratics. We prove that our model-based procedure converges in noisy gradient setting. This is an interesting step for constructing self-tuning quadratics.
  arXiv  Detail & Related papers  (2020-11-09T22:07:30Z)
• A Primal-Dual Subgradient Approachfor Fair Meta Learning [23.65344558042896]
  Few shot meta-learning is well-known with its fast-adapted capability and accuracy generalization onto unseen tasks. We propose a Primal-Dual Fair Meta-learning framework, namely PDFM, which learns to train fair machine learning models using only a few examples.
  arXiv  Detail & Related papers  (2020-09-26T19:47:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.