Kun Liu

Kun Liu

CS & Applied Math & MPC @ GZU | Phd

A Review about Privacy-preserving Machine Learning

1 minute read

Published:

A basic protocol reference about privacy preserving machine learning.

Improvement Derivation
Optimization Low latency privacy preserving inference
  GAZELLE: A low latency framework for secure neural network inference
Matrix Multiplication Secure outsourced matrix computation and application to neural networks
  More practical privacy- preserving machine learning as A service via efficient secure matrix multiplication
Non-linear Function Improved Primitives for MPC over Mixed Arithmetic-Binary Circuits
Linear Function nGraph-HE: a graph compiler for deep learning on homomorphically encrypted data
  CHET: an optimizing compiler for fully-homomorphic neural-network inferencing
  Privacy-preserving machine learning as a service
Binary Function QUOTIENT: Two-Party Secure Neural Network Training and Prediction
  XONN: XNOR-based oblivious deep neural network inference
Mixed crpto protocol Secure evaluation of quantized neural networks
  GAZELLE: A low latency framework for secure neural network inference
  Oblivious neural network predictions via minionn transformations
  SecureML: A system for scalable privacy-preserving machine learning
  ABY2.0: improved mixed-protocol secure two-party computation
  Slalom: Fast, verifiable and private execution of neural networks in trusted hardware
Non-HE Secure evaluation of quantized neural networks
  Fantastic four: Honest-majority four-party secure computation with malicious security
  SWIFT: super-fast and robust privacy-preserving machine learning
  CrypT- Flow: Secure tensorflow inference
  ABY3: A Mixed Protocol Framework for Machine Learning
HE DELPHI: A cryptographic inference service for neural networks
  CrypTFlow2: Practical 2-party secure inference
  SortingHat: Efficient Private Decision Tree Evaluation via Homomorphic Encryption and Transciphering
RLHE Cheetah: Lean and Fast Secure Two-Party Deep Neural Network Inference

You May Also Enjoy