Making AI Forget You: Data Deletion in Machine Learning

By Antonio A. Ginart, Melody Y. Guan, Gregory Valiant, James Y. Zou. 2019

709 citations recorded by the source corpus.

Intense recent discussions have focused on how to provide individuals with control over when their data can and cannot be used --- the EU's Right To Be Forgotten regulation is an example of this effort. In this paper we initiate a framework studying what to do when it is no longer permissible to deploy models derivative from specific user data. In particular, we formulate the problem of efficiently deleting individual data points from trained machine learning models. For many standard ML models, the only way to completely remove an individual's data is to retrain the whole model from scratch on the remaining data, which is often not computationally practical. We investigate algorithmic principles that enable efficient data deletion in ML. For the specific setting of k-means clustering, we propose two provably efficient deletion algorithms which achieve an average of over 100X improvement in deletion efficiency across 6 datasets, while producing clusters of comparable statistical quality to a canonical k-means++ baseline.

Subfields: Artificial Intelligence, Management Science and Operations Research, Privacy & Security, AI Ethics, Fairness & Society

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Management Science and Operations Research

In Privacy & Security

In AI Ethics, Fairness & Society

Citations in the corpus

Secure Federated Submodel Learning (2019, background): “or company is prohibited from uploading user data and storing it in the cloud without the explicit permissions from the European Union users [7], [8].”
Learning with Selective Forgetting (2021, methodology): “Specialized methods for some basic learning algorithms such as linear discriminant analysis [Guo et al. , 2020] and k-means [Ginart et al. , 2019] have also been presented.”
Mixed-Privacy Forgetting in Deep Networks (2020, methodology): “Can we forget the cohort of interest without interfering with information about the other data and preserving, to the extent possible, the accuracy of the trained model? Recently, the problem of forgetting has received considerable attention [15, 16, 13, 19, 5, 24, 35, 41, 6, 39, 12, 7, 36], but sol”
DeltaGrad: Rapid retraining of machine learning models (2020, background): “Such ideas have been used recently e.g., Ginart et al. (2019); Guo et al. (2019); Wu et al. (2020).”
Machine Unlearning (2019, background): “Decremental Learning: Ginart et al. [37] consider the problem from a data-protection regulation standpoint.”
Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networks (2019, methodology): “[11] proposed an efficient data elimination algorithm for k-means clustering.”
Measuring Forgetting of Memorized Training Examples (2022, background): “Algorithms for machine unlearning have been proposed for k-means [24], empirical risk minimization [29, 35, 59, 76, 67], and deep learning [18, 26, 25, 60, 5].”
Certified Data Removal from Machine Learning Models (2019, background): “Ginart et al. (2019) studied the problem of removing data from k -means clusterings.”
A Unified and General Framework for Continual Learning (2024, background): “Machine Unlearning (Guo et al., 2020; Wu et al., 2020; Bourtoule et al., 2021; Ullah et al., 2021) refers to the process of removing or erasing previously learned information or knowledge from a pre-trained model to comply with privacy regulations (Ginart et al., 2019).”
SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation (2023, background): “MU aims at modifying ML models to eliminate the influence of specific data points or classes, initially developed to mitigate potential privacy breaches post-training (Ginart et al., 2019; Neel et al., 2021; Ullah et al., 2021; Sekhari et al., 2021).”
Detecting Pretraining Data from Large Language Models (2023, methodology): “…unlearning has emerged as a solution for purging data from machine learning models, and various machine unlearning methods have been introduced (Ginart et al., 2019; Liu et al., 2020; Wu et al., 2020; Bourtoule et al., 2021; Izzo et al., 2021; Sekhari et al., 2021; Gupta et al., 2021; Ye et…”
A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection (2019, background): “user wants to delete its data, should the global model be retrained without the data [58]? There is still a gap between the FL techniques and the regulations in reality.”
Machine Unlearning for Random Forests (2020, background): “Ginart et al. (2019) introduced a quantized variant of the k -means algo-rithm (Lloyd, 1982) called Q-k -means that supports exact data deletion.”
Variational Bayesian Unlearning (2020, methodology): “, naive Bayes classifier, linear regression, k-means clustering, and item-item collaborative filtering [4, 12, 30]).”
Remember What You Want to Forget: Algorithms for Machine Unlearning (2021, methodology): “This requirement is rather strong: Ginart et al. [2019] proposed a relaxed notion of deletion, in which the model must only be close to the alternative, where closeness is defined in a way reminiscent of differential privacy [Dwork et al.”
Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning (2024, methodology): “Since its proposal by Cao & Yang (2015), machine unlearning has been extensively studied in the classification literature (Bourtoule et al., 2021; Golatkar et al., 2020; Ginart et al., 2019; Thudi et al., 2022; Izzo et al., 2021; Koh & Liang, 2017; Guo et al., 2019; Sekhari et al., 2021).”
DART: Data Addition and Removal Trees (2020, neutral): “As previous work has noted, differential privacy is a sufﬁ-cient condition for efﬁcient data deletion, but not a necessary one (Ginart et al. 2019).”
Knowledge Unlearning for Mitigating Privacy Risks in Language Models (2022, background): “2.2 Machine Unlearning Machine unlearning has received attention as an alternative approach to overcome data privacy issues in machine learning (Cao and Yang, 2015; Ginart et al., 2019; Bourtoule et al., 2021; Graves et al., 2021).”
Forgetting Outside the Box: Scrubbing Deep Networks of Information Accessible from Input-Output Observations (2020, background): “[14] formalizes the problem of efficient data elimination, and provides engineering principles for designing forgetting algorithms.”