FedGA-Tree: Federated Decision Tree using Genetic Algorithm
This addresses the need for nonparametric models in federated learning, offering a personalized approach for clients with privacy concerns, though it is incremental as it builds on existing genetic algorithm and federated learning techniques.
The paper tackles the problem of adapting decision trees to federated learning by proposing FedGA-Tree, which uses a genetic algorithm to construct personalized decision trees for both classification and regression with categorical and numerical data. The method outperforms local training and a benchmark algorithm in experiments.
In recent years, with rising concerns for data privacy, Federated Learning has gained prominence, as it enables collaborative training without the aggregation of raw data from participating clients. However, much of the current focus has been on parametric gradient-based models, while nonparametric counterparts such as decision tree are relatively understudied. Existing methods for adapting decision trees to Federated Learning generally combine a greedy tree-building algorithm with differential privacy to produce a global model for all clients. These methods are limited to classification trees and categorical data due to the constraints of differential privacy. In this paper, we explore an alternative approach that utilizes Genetic Algorithm to facilitate the construction of personalized decision trees and accommodate categorical and numerical data, thus allowing for both classification and regression trees. Comprehensive experiments demonstrate that our method surpasses decision trees trained solely on local data and a benchmark algorithm.