Online Gradient Boosting Decision Tree: In-Place Updates for Efficient Adding/Deleting Data
This work addresses the need for dynamic data updates in GBDT models, which is incremental as it extends existing methods to support online modifications.
The paper tackles the problem of traditional Gradient Boosting Decision Tree (GBDT) models not supporting data addition or deletion after training by proposing an efficient online learning framework that enables both incremental and decremental learning. The results include theoretical analysis of hyper-parameters for accuracy-cost trade-offs and empirical validation on public datasets, with backdoor attack experiments showing successful injection and removal of backdoors.
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.