TAPAS: Tricks to Accelerate (encrypted) Prediction As a Service

By Amartya Sanyal, Matt J. Kusner, Adrià Gascón, Varun Kanade. 2018

135 citations recorded by the source corpus.

Machine learning methods are widely used for a variety of prediction problems. \emph{Prediction as a service} is a paradigm in which service providers with technological expertise and computational resources may perform predictions for clients. However, data privacy severely restricts the applicability of such services, unless measures to keep client data private (even from the service provider) are designed. Equally important is to minimize the amount of computation and communication required between client and server. Fully homomorphic encryption offers a possible way out, whereby clients may encrypt their data, and on which the server may perform arithmetic computations. The main drawback of using fully homomorphic encryption is the amount of time required to evaluate large machine learning models on encrypted data. We combine ideas from the machine learning literature, particularly work on binarization and sparsification of neural networks, together with algorithmic tools to speed-up and parallelize computation using encrypted data.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

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

Citations in the corpus

Secure Evaluation of Quantized Neural Networks (2019, methodology): “Sanyal et al. [57] use the same techniques.”
QUOTIENT: Two-Party Secure Neural Network Training and Prediction (2019, methodology): “While there have been massive practical developments in both cryptography and ML (including the works above), most recent works for model training [22, 40, 41, 54] and prediction [7, 15, 50] on encrypted data are largely based on optimizations for either the ML model or the employed cryptographic te”
PrivEdge: From Local to Distributed Private Training and Prediction (2020, background): “CryptoNets [38] and TAPAS [39] do private prediction using homomorphic encryption.”
Security and Privacy Issues in Deep Learning (2018, methodology): “In comparison, TAPAS [20] binarized weights and enabled binary operations and sparsifications techniques.”
XONN: XNOR-based Oblivious Deep Neural Network Inference (2019, result): “Concurrently and independently of ours, in TAPAS [53], Sanyal et al. study the binarization of neural networks in the context of oblivious inference.”
Cheetah: Optimizing and Accelerating Homomorphic Encryption for Private Inference (2020, methodology): “Techniques can be categorized into two groups: HE only [11], [29], [32], [55], or multiparty computation (MPC)- based [33], [37], [50], [51].”
A Hybrid Deep Learning Architecture for Privacy-Preserving Mobile Analytics (2017, methodology): “computational complexity of encryption-based methods using parallelization techniques, but their method still requires more than 100 s to be run on a 16-machine cluster [48].”
Low Latency Privacy Preserving Inference (2018, methodology): “, 2016) was the first demonstration of a privacy preserving Encrypted Prediction as a Service (EPaaS) solution (Sanyal et al., 2018) based on HE.”