Federated Gradient Boosting

Federated gradient boosting aims to train powerful gradient boosting machine (GBM) models on decentralized data without directly sharing sensitive information, addressing privacy concerns in collaborative machine learning. Current research focuses on improving efficiency and privacy through novel algorithms like gradient-less approaches, adapting GBMs to hybrid and vertical data settings, and incorporating differential privacy mechanisms into the training process. These advancements are significant for enabling secure and efficient model training across multiple parties, with applications ranging from healthcare to finance where data sharing is restricted.

Papers

November 22, 2023

SecureCut: Federated Gradient Boosting Decision Trees with Efficient Machine Unlearning
Jian Zhang, Bowen Li Jie Li, Chentao Wu
Machine Unlearning Unlearning Framework Vertical Federated Learning Cut and Approximate Gradient Boosting Decision Tree Data Deletion Federated Gradient Boosting

October 18, 2023

Effective and Efficient Federated Tree Learning on Hybrid Data
Qinbin Li, Chulin Xie, Xiaojun Xu, Xiaoyuan Liu, Ce Zhang, Bo Li, Bingsheng He, Dawn Song
Hybrid Data Federated Gradient Boosting Balanced Split

April 15, 2023

Gradient-less Federated Gradient Boosting Trees with Learnable Learning Rates
Chenyang Ma, Xinchi Qiu, Daniel J. Beutel, Nicholas D. Lane
Learning Rate Decentralized Data XGBoost Model Tree Ensemble Extreme Gradient Boosting Federated Gradient Boosting

October 6, 2022

Federated Boosted Decision Trees with Differential Privacy
Samuel Maddock, Graham Cormode, Tianhao Wang, Carsten Maple, Somesh Jha
Differential Privacy Homomorphic Encryption Privacy Preserving Machine Learning Gradient Boosted Decision Tree Federated Gradient Boosting

October 4, 2022

OpBoost: A Vertical Federated Tree Boosting Framework Based on Order-Preserving Desensitization
Xiaochen Li, Yuke Hu, Weiran Liu, Hanwen Feng, Li Peng, Yuan Hong, Kui Ren, Zhan Qin
Differential Privacy Privacy Preserving Vertical Federated Learning Local Differential Privacy Image Desensitization Federated Gradient Boosting

April 3, 2022

FedGBF: An efficient vertical federated learning framework via gradient boosting and bagging
Yujin Han, Pan Du, Kai Yang
Decision Tree Gradient Boosting Learning Framework Boosting Algorithm Federated Gradient Boosting