$\nabla \tau$: Gradient-based and Task-Agnostic machine Unlearning

arXiv:2403.14339v1 Announce Type: new
Abstract: Machine Unlearning, the process of selectively eliminating the influence of certain data examples used during a model's training, has gained significant attention as a means for practitioners to comply with recent data protection regulations. However, existing unlearning methods face critical drawbacks, including their prohibitively high cost, often associated with a large number of hyperparameters, and the limitation of forgetting only relatively small data portions. This often makes retraining the model from scratch a quicker and …

abstract arxiv attention cost cs.ai cs.cl cs.cv cs.lg data data protection examples face gradient however influence machine nabla process protection regulations training type unlearning