Efficient Skip Connections Realization for Secure Inference on Encrypted Data

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Drucker, Nir, Zimerman, Itamar
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866929572437557248
author Drucker, Nir
Zimerman, Itamar
author_facet Drucker, Nir
Zimerman, Itamar
contents Homomorphic Encryption (HE) is a cryptographic tool that allows performing computation under encryption, which is used by many privacy-preserving machine learning solutions, for example, to perform secure classification. Modern deep learning applications yield good performance for example in image processing tasks benchmarks by including many skip connections. The latter appears to be very costly when attempting to execute model inference under HE. In this paper, we show that by replacing (mid-term) skip connections with (short-term) Dirac parameterization and (long-term) shared-source skip connection we were able to reduce the skip connections burden for HE-based solutions, achieving x1.3 computing power improvement for the same accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2306_06736
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Efficient Skip Connections Realization for Secure Inference on Encrypted Data
Drucker, Nir
Zimerman, Itamar
Cryptography and Security
Machine Learning
Homomorphic Encryption (HE) is a cryptographic tool that allows performing computation under encryption, which is used by many privacy-preserving machine learning solutions, for example, to perform secure classification. Modern deep learning applications yield good performance for example in image processing tasks benchmarks by including many skip connections. The latter appears to be very costly when attempting to execute model inference under HE. In this paper, we show that by replacing (mid-term) skip connections with (short-term) Dirac parameterization and (long-term) shared-source skip connection we were able to reduce the skip connections burden for HE-based solutions, achieving x1.3 computing power improvement for the same accuracy.
title Efficient Skip Connections Realization for Secure Inference on Encrypted Data
topic Cryptography and Security
Machine Learning
url https://arxiv.org/abs/2306.06736