CRAIJan 29, 2024

Neural Network Training on Encrypted Data with TFHE

arXiv:2401.16136v14 citationsh-index: 7
Originality Synthesis-oriented
AI Analysis

This enables secure outsourcing and collaboration on confidential data for parties with privacy concerns, though it is incremental as it applies existing encryption methods to neural network training.

The paper tackles the problem of training neural networks on encrypted data to preserve confidentiality, achieving results by training logistic regression and multi-layer perceptrons on several datasets using fully homomorphic encryption.

We present an approach to outsourcing of training neural networks while preserving data confidentiality from malicious parties. We use fully homomorphic encryption to build a unified training approach that works on encrypted data and learns quantized neural network models. The data can be horizontally or vertically split between multiple parties, enabling collaboration on confidential data. We train logistic regression and multi-layer perceptrons on several datasets.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes