Differentially private recommender systems: building privacy into the net

By Frank McSherry, Ilya Mironov. 2009

825 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Sociology and Political Science, Recommender Systems, 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 Sociology and Political Science

In Recommender Systems

In Privacy & Security

Citations in the corpus

Deep Models Under the GAN: Information Leakage from Collaborative Deep Learning (2017, background): “Several works have adopted DP as an efficient defense mechanism such as [5, 7, 11, 12, 18, 23, 24, 37, 53, 60, 65, 72, 86].”
Certified Robustness to Adversarial Examples with Differential Privacy (2018, methodology): “Significant work has focused on making ML algorithms DP to preserve the privacy of training sets [39], [1], [9].”
Functional Mechanism: Regression Analysis under Differential Privacy (2012, methodology): “…[2] and Ding et al. [8] consider the publication of data cubes; Xu et al. [31] and Li et al. [18] focus on publishing histograms; McSherry and Mironov [20], Rastogi and Nath [24], and McSherry and Mahajan [19] devise methods for re-leasing counts on particular types of data, such as time series.”
Learning in a Large Function Space: Privacy-Preserving Mechanisms for SVM Learning (2009, methodology): “…interval queries and half-spaces as explored by Blum et al. 2008), or on constructing learning algorithms that can be decomposed into subset-sum operations (e.g., perceptron, k-NN, ID3 as described by Blum et al. 2005, and various recommender systems due to the work of McSherry and Mironov 2009).”
Building a RAPPOR with the Unknown: Privacy-Preserving Learning of Associations and Data Dictionaries (2015, background): “Research on local differential privacy has largely been centered around finding algorithms that satisfy differential privacy properties [19, 21, 22], and improving the tradeoffs between privacy and utility [7, 14].”
Learning Differentially Private Language Models Without Losing Accuracy (2017, background): “The main goal of this work is to provide a strong guarantee that the trained model protects the privacy of individuals’ data without undue sacriﬁce in model quality, building upon recent work in applying differential privacy to machine learning [11, 10, 16, 8, 6, 5, 18, 24, 22, 1, 20, 26].”
Towards Making Systems Forget with Machine Unlearning (2015, background): “[52] built a differentially private recommendation system and showed”
Efficient Private Statistics with Succinct Sketches (2015, methodology): “McSherry and Mironov [51] propose a privacy-preserving recommender system that relies on trusted computing, while Ciss´ee and Albayrak [20] use differential privacy to add privacy guarantees to a few algorithms presented during the Netﬂix Prize competition.”
Privacy via the Johnson-Lindenstrauss Transform (2012, methodology): “[17], which has gained widespread recognition in recent years (see a survey [15]), and has been used to demonstrate the feasibility of privacy-preserving data releases [3, 27, 33, 34].”
The Johnson-Lindenstrauss Transform Itself Preserves Differential Privacy (2012, methodology): “The work of McSherry and Mironov [24] (inspired by the Netflix prize competition) defines neighboring databases as a change in a single entry, and introduces O(k(2)) noise while outputting a rank-k approximation of the input.”
Differentially-private network trace analysis (2010, background): “Appealingly, it is resilient to collusion, supports multiple interactive queries, and is also independent of any auxiliary information that an attacker might possess; such information has been shown to break anonymization [5, 26, 20, 29].”
The matrix mechanism: optimizing linear counting queries under differential privacy (2015, background): “Approximate ( , δ)-differential privacy can be achieved by adding Gaussian noise calibrated to the L2 sensitivity of the queries [10,22].”
Evaluating Differentially Private Machine Learning in Practice (2019, background): “Some of the early data analytics works on frequent pattern mining [7,41], decision trees [21], private record linkage [30] and recommender systems [48] were able to achieve both high utility and privacy with settings close to 1.”
When Relaxations Go Bad: "Differentially-Private" Machine Learning (2019, background): “Some of the early data analytics works on frequent pattern mining [6,41], decision trees [22], private record linkage [31] and recommender systems [47] were able to achieve both high utility and privacy with settings close to 1.”
On differentially private low rank approximation (2013, methodology): “The work of McSherry and Mironov[12] applied input perturbation techniques to the Netflix dataset.”
A Simple and Efficient Federated Recommender System (2019, methodology): “Among the extensive research to secure privacy in recommender systems are methods that include cryptography algorithms [7, 23] and differential privacy [4, 20].”