Train a Convolutional Neural Nets Using Multi-Party Computation
Most efforts to privately train machine-learning models either use homomorphic encryption or secure multi-party computation (MPC). Recent insights indicate…
Train a Convolutional Neural Nets Using Multi-Party Computation launched on June 26, 2018 with 28 votes, #759 of 2,653 launches that month and more than 75% of that year's launches. The site was still online in September 2026, 8 years on.
In plain words
This project demonstrates how to train convolutional neural networks using secure multi-party computation (MPC), specifically the SPDZ protocol. It addresses private machine learning by enabling neural network training without exposing sensitive data to a single party. The implementation builds on previous work in privately fine-tuning CNNs and is designed for developers and researchers working on privacy-preserving machine learning applications.
written from the facts on this page · September 2026
From the sources
Most efforts to privately train machine-learning models either use homomorphic encryption or secure multi-party computation (MPC). Recent insights indicate that the latter has the most potential. Here we will describe the implementation of Convolutional Neural Networks (CNNs) using MPC. This post is heavily inspired by earlier work that describes how CNNs can be privately fine-tuned after first training on a public dataset. As this post builds upon concepts in that previous post, it might be wise to take a look at that first. The focus of this post is slightly different: What if we want to train a CNN from scratch without transfer learning? This requires some complex private operations with…from blog.openmined.org
Does a similar job
all alternatives →- IMI made a neural net that analyzes privacy policies2019 · useguard.com · ▲502
Thinc, a new deep learning library by the makers of spaCy and FastAPI2020 · thinc.ai · ▲250Thinc is a lightweight type-checked deep learning library for composing models, with support for layers defined in frameworks like PyTorch and TensorFlow.
A perceptron with stochastic gradient descent in Go2017 · github.com · ▲94A single / multi layer / recurrent neural network written in Golang. - made2591/go-perceptron-go
ArcaneVM – A Fully Homomorphic Encryption Brainfuck Virtual Machine2019 · github.com · ▲131A Fully Homomorphic Encryption Brainfuck virtual machine - f-prime/arcanevm
Encrypted Deep Learning Classification with PyTorch and PySyft2019 · blog.openmined.org · ▲10Update as of November 18, 2021: The version of PySyft mentioned in this post has been deprecated. Any implementations using this older version of PySyft are…
Multi-GPU Reinforcement Learning in Tensorflow for OpenAI Gym2016 · github.com · ▲58Multi-GPU reinforcement learning using Deep Q-Network in TensorFlow for OpenAI Gym - vishar0/dist-dqn
More ai this month
the category →
Paintings by AI models: programs run through a simulation of oil paint.
AI · 21d ago · stillwet.art
PixVerse R2▲381A real-time world model you can explore and change
AI · 29d ago · world.pixverse.video
Source: Show HN post ↗
Launched alongside, June 2018
the whole month →
- OL
Dev tools · 2018 · openlogos.org
Google Podcasts▲648Discover and listen to the world's podcasts on Android.
Work · 2018 · play.google.com
This quick quiz helps you understand why you
Life & fun · 2018 · whydoiprocrastinate.com