Online Gradient Boosting Decision Tree: In-Place Updates for Efficient Adding/Deleting Data

TL;DR

This paper introduces a novel online GBDT framework that supports efficient in-place incremental and decremental learning, enabling dynamic data updates without retraining from scratch, with theoretical and empirical validation.

Contribution

It presents the first unified in-place incremental and decremental learning framework for GBDT, with optimizations and theoretical analysis for balancing accuracy and computational cost.

Findings

Framework effectively supports data addition and deletion on the fly.

Theoretical analysis guides hyper-parameter tuning for accuracy-cost trade-offs.

Empirical results demonstrate efficiency and robustness in real datasets.

Abstract

Gradient Boosting Decision Tree (GBDT) is one of the most popular machine learning models in various applications. However, in the traditional settings, all data should be simultaneously accessed in the training procedure: it does not allow to add or delete any data instances after training. In this paper, we propose an efficient online learning framework for GBDT supporting both incremental and decremental learning. To the best of our knowledge, this is the first work that considers an in-place unified incremental and decremental learning on GBDT. To reduce the learning cost, we present a collection of optimizations for our framework, so that it can add or delete a small fraction of data on the fly. We theoretically show the relationship between the hyper-parameters of the proposed optimizations, which enables trading off accuracy and cost on incremental and decremental learning. The backdoor attack results show that our framework can successfully inject and remove backdoor in a well-trained model using incremental and decremental learning, and the empirical results on public datasets confirm the effectiveness and efficiency of our proposed online learning framework and optimizations.