HeLayers: A Tile Tensors Framework for Large Neural Networks on Encrypted Data
This work addresses privacy-preserving machine learning for companies needing to comply with regulations by enabling faster encrypted computations, though it is incremental as it builds on existing HE methods.
The paper tackles the challenge of efficiently running large neural networks on encrypted data using Homomorphic Encryption (HE) by introducing HeLayers, a framework that abstracts data packing decisions and includes a novel algorithm for 2D convolutions. The result is an HE-friendly version of AlexNet that runs in three minutes, significantly faster than prior HE-only solutions.
Privacy-preserving solutions enable companies to offload confidential data to third-party services while fulfilling their government regulations. To accomplish this, they leverage various cryptographic techniques such as Homomorphic Encryption (HE), which allows performing computation on encrypted data. Most HE schemes work in a SIMD fashion, and the data packing method can dramatically affect the running time and memory costs. Finding a packing method that leads to an optimal performant implementation is a hard task. We present a simple and intuitive framework that abstracts the packing decision for the user. We explain its underlying data structures and optimizer, and propose a novel algorithm for performing 2D convolution operations. We used this framework to implement an HE-friendly version of AlexNet, which runs in three minutes, several orders of magnitude faster than other state-of-the-art solutions that only use HE.