Mixed-Privacy Forgetting in Deep Networks

By Aditya Golatkar, Alessandro Achille, Avinash Ravichandran, Marzia Polito, Stefano Soatto. 2020

233 citations recorded by the source corpus.

We show that the influence of a subset of the training samples can be removed – or "forgotten" – from the weights of a network trained on large-scale image classification tasks, and we provide strong computable bounds on the amount of remaining information after forgetting. Inspired by real-world applications of forgetting techniques, we introduce a novel notion of forgetting in mixed-privacy setting, where we know that a "core" subset of the training samples does not need to be forgotten. While this variation of the problem is conceptually simple, we show that working in this setting significantly improves the accuracy and guarantees of forgetting methods applied to vision classification tasks. Moreover, our method allows efficient removal of all information contained in non-core data by simply setting to zero a subset of the weights with minimal loss in performance. We achieve these results by replacing a standard deep network with a suitable linear approximation. With opportune changes to the network architecture and training procedure, we show that such linear approximation achieves comparable performance to the original network and that the forgetting problem becomes quadratic and can be solved efficiently even for large models. Unlike previous forgetting methods on deep networks, ours can achieve close to the state-of-the-art accuracy on large scale vision tasks. In particular, we show that our method allows forgetting without having to trade off the model accuracy.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Privacy & Security

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 Computer Vision and Pattern Recognition

In Privacy & Security

Citations in the corpus

A Comprehensive Survey of Continual Learning: Theory, Method and Application (2023, neutral): “…oftrainingsamples[307],utilizinghistoricalparametersand gradients[308],removingoldknowledgefromparameterswith Fisherinformationmatrix[309],addingadaptiveparameterson apre-trainedmodel[310],etc. Ontheotherhand,continual learningmaysufferfromdataleakageandprivacyinvasionas retainingalloldknowledge.”
Learning with Selective Forgetting (2021, background): “Mixed-Linear Forgetting [Golatkar et al. , 2021] derived a tractable optimization problem by linearly approximating the amount of change in weights due to the addition of training data.”
Measuring Forgetting of Memorized Training Examples (2022, background): “Algorithms for machine unlearning have been proposed for k -means [26], empirical risk minimization [31, 37, 59, 77, 68], and deep learning [20, 28, 27, 60, 5].”
Ablating Concepts in Text-to-Image Diffusion Models (2023, background): “Another line of work in machine unlearning [9, 21, 23, 22, 42, 8, 67, 60] explores data deletion at user’s request after model training.”
Fast Yet Effective Machine Unlearning (2021, background): “Some privacy settings [18, 26] allow the user to make a request to forget their data from the already trained model.”
Adaptive Machine Unlearning (2021, methodology): “It also includes the approach of Golatkar et al. [2020a,b] which pre-trains a nonconvex model on data that will never be deleted and then does convex ﬁne-tuning on user data on top of that.”
Machine Unlearning of Features and Labels (2021, methodology): “[25, 26] use influence functions for sample removal in deep neural networks by proposing special approximations of the learning model.”
Machine Unlearning for Random Forests (2020, background): “…2019), SVMs (Cauwenberghs & Poggio, 2001), logistic regression (Guo et al., 2020; Schelter, 2020), and deep neural networks (Baumhauer et al., 2020; Golatkar et al., 2020b; Wu et al., 2020); however, there is very limited work addressing the problem of efﬁcient data deletion for tree-based models…”
Continual Learning and Private Unlearning (2022, background): “Mixed-Linear Forgetting proposes a tractable optimization problem by lineary approximating the amount of change in weights due to the addition of any training data (Golatkar et al., 2021).”
Remember What You Want to Forget: Algorithms for Machine Unlearning (2021, background): “Finally, there has also been recent empirical and theoretical work in developing deﬁnitions and algorithms for machine unlearning with deep neural networks for application domains in computer vision (Du et al., 2019; Golatkar et al., 2020b,a; Nguyen et al., 2020).”
Towards Unbounded Machine Unlearning (2023, background): “[ Golatkar et al., 2021] addresses this issue, albeit under a different restrictive assumption: that a large dataset is available for pre-training that will remain ‘static’, in that no forgetting operations will be applied on it; an assumption that is not met in many important applications (e.g.…”