loretanr/dp-gbdt
GBDT learning + differential privacy. Standalone C++ implementation of "DPBoost" (Li et al.). There are further hardened & SGX versions of the code.
This project provides an implementation of differentially private Gradient Boosted Decision Trees (GBDT). It takes structured data as input and produces a GBDT model that can perform regression or binary classification while protecting the privacy of individual data points. This is ideal for data scientists, machine learning engineers, or researchers working with sensitive datasets who need to build predictive models without compromising privacy.
No commits in the last 6 months.
Use this if you need to build machine learning models with strong privacy guarantees, especially when dealing with sensitive data that requires differential privacy.
Not ideal if you need to perform multi-class classification, or if your primary concern is model performance without the specific requirement for differential privacy, as there are still minor privacy issues being resolved.
Stars
9
Forks
1
Language
C++
License
MIT
Category
Last pushed
May 19, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/loretanr/dp-gbdt"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
tensorflow/privacy
Library for training machine learning models with privacy for training data
meta-pytorch/opacus
Training PyTorch models with differential privacy
tf-encrypted/tf-encrypted
A Framework for Encrypted Machine Learning in TensorFlow
awslabs/fast-differential-privacy
Fast, memory-efficient, scalable optimization of deep learning with differential privacy
privacytrustlab/ml_privacy_meter
Privacy Meter: An open-source library to audit data privacy in statistical and machine learning...