Secure and Scalable Document Similarity on Distributed Databases: Differential Privacy to the Rescue

Schoppmann, Phillipp; Vogelsang, Lennart; Gascón, Adrià; Balle, Borja

doi:10.2478/popets-2020-0024

Cited by 5 publications

(4 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Securely computing kNNC is another closely related field when data is stored in different local devices. Majority of the frameworks that ensure privacy for kNNC often use some sort of secure multi-party computation (SMC) protocols (Zhan, Chang, and Matwin 2008;Xiong, Chitti, and Liu 2006;Qi and Atallah 2008;Schoppmann et al 2020;Shaul, Feldman, and Rus 2020;Chen et al 2020).…”

Section: Related Workmentioning

confidence: 99%

“…In most secure kNNC settings considered in the literature, the goal is to keep the training data secure from the party making the test query (Qi and Atallah 2008;Shaul, Feldman, and Rus 2018;Wu et al 2019) and it is not clear how those approaches extend to the multiparty federated setting where the per-party data (train or test) should remain localized. Of particular relevance is Schoppmann et al (2020) which proposed a scheme to compute a secure inner-product between any test point and all training points (distributed across parties) and then perform a secure top-k protocol to perform kNNC. This procedure explicitly computes the neighbors for a test point, which involves n secure similarity computations for each test point (on top of the secure top-k protocol).…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Federated Nearest Neighbor Classification with a Colony of Fruit-Flies

Ram¹,

Sinha

2022

AAAI

View full text Add to dashboard Cite

The mathematical formalization of a neurological mechanism in the fruit-fly olfactory circuit as a locality sensitive hash (Flyhash) and bloom filter (FBF) has been recently proposed and "reprogrammed" for various learning tasks such as similarity search, outlier detection and text embeddings. We propose a novel reprogramming of this hash and bloom filter to emulate the canonical nearest neighbor classifier (NNC) in the challenging Federated Learning (FL) setup where training and test data are spread across parties and no data can leave their respective parties. Specifically, we utilize Flyhash and FBF to create the FlyNN classifier, and theoretically establish conditions where FlyNN matches NNC. We show how FlyNN is trained exactly in a FL setup with low communication overhead to produce FlyNNFL, and how it can be differentially private. Empirically, we demonstrate that (i) FlyNN matches NNC accuracy across 70 OpenML datasets, (ii) FlyNNFL training is highly scalable with low communication overhead, providing up to 8x speedup with 16 parties.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Federated Nearest Neighbor Classification with a Colony of Fruit-Flies

Ram¹,

Sinha

2022

AAAI

View full text Add to dashboard Cite

show abstract

“…Recently, Schoppmann et al [21] have proposed a secure scheme for classifying collaborated documents, where a k-nearest neighbor classifier is applied to compute the secure similarity of textual documents.…”

Section: Related Workmentioning

confidence: 99%

Privacy Preserving Scheme for Document Similarity Detection

2021

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

The problem of detecting similar documents plays an essential role for many real-world applications, such as copyright protection and plagiarism detection. To protect data privacy, the new version of such a problem becomes more challenging, where the matched documents are distributed among two or more parties and their privacy should be preserved. In this paper, we propose new privacy-preserving document similarity detection schemes by utilizing the locality-sensitive hashing technique, which can handle the misspelled mistakes. Furthermore, the keywords' occurrences of a given document are integrated into its underlying representation to support a better ranking for the returned results.We introduced a new security definition, which hides the exact similarity scores towards the querying party. Extensive experiments on real-world data illustrate that our proposed schemes are efficient and accurate.

show abstract

“…A previous version ofSchoppmann et al (2020) was titled "Private Nearest Neighbors Classification in Federated Databases" (https://eprint.iacr.org/eprint-bin/versions.pl?entry=2018/289) but has since been changed as the focus of the paper was shifted.…”

mentioning

confidence: 99%

Federated Nearest Neighbor Classification with a Colony of Fruit-Flies: With Supplement

Ram¹,

Sinha²

2021

Preprint

View full text Add to dashboard Cite

The mathematical formalization of a neurological mechanism in the olfactory circuit of a fruit-fly as a locality sensitive hash (FlyHash) and bloom filter (FBF) has been recently proposed and "reprogrammed" for various machine learning tasks such as similarity search, outlier detection and text embeddings. We propose a novel reprogramming of this hash and bloom filter to emulate the canonical nearest neighbor classifier (NNC) in the challenging Federated Learning (FL) setup where training and test data are spread across parties and no data can leave their respective parties. Specifically, we utilize FlyHash and FBF to create the FlyNN classifier, and theoretically establish conditions where FlyNN matches NNC. We show how FlyNN is trained exactly in a FL setup with low communication overhead to produce FlyNNFL, and how it can be differentially private. Empirically, we demonstrate that (i) FlyNN matches NNC accuracy across 70 OpenML datasets, (ii) FlyNNFL training is highly scalable with low communication overhead, providing up to 8× speedup with 16 parties.

show abstract

Secure and Scalable Document Similarity on Distributed Databases: Differential Privacy to the Rescue

Cited by 5 publications

References 41 publications

Federated Nearest Neighbor Classification with a Colony of Fruit-Flies

Federated Nearest Neighbor Classification with a Colony of Fruit-Flies

Privacy Preserving Scheme for Document Similarity Detection

Federated Nearest Neighbor Classification with a Colony of Fruit-Flies: With Supplement

Contact Info

Product

Resources

About