Neural Network Training on Encrypted Data with TFHE
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916109223985152 |
|---|---|
| author | Montero, Luis Frery, Jordan Kherfallah, Celia Bredehoft, Roman Stoian, Andrei |
| author_facet | Montero, Luis Frery, Jordan Kherfallah, Celia Bredehoft, Roman Stoian, Andrei |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_16136 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Neural Network Training on Encrypted Data with TFHE Montero, Luis Frery, Jordan Kherfallah, Celia Bredehoft, Roman Stoian, Andrei Cryptography and Security Artificial Intelligence 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. |
| title | Neural Network Training on Encrypted Data with TFHE |
| topic | Cryptography and Security Artificial Intelligence |
| url | https://arxiv.org/abs/2401.16136 |