DeltaBoost: Gradient Boosting Decision Trees with Efficient Machine Unlearning

Summary: DeltaBoost tailors GBDT for efficient unlearning, enabling deletion of specific records with preserved utility. Robust GBDT-like design with decoupled training minimizes inter-tree dependence, delivering up to 100x speedup over retraining on five datasets. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID: h907a4405c42fee9c
Venue: SIGMOD
Year: 2023
Pagerank: 5.5025409e-05
Overall Rank: 7,524 | 49.42%
DOI: 10.1145/3589313

Incoming Non-self Citations Over Time

Authors

1. Zhaomin Wu (National University of Singapore)
2. Junhui Zhu (National University of Singapore)
3. Qinbin Li (National University of Singapore)
4. Bingsheng He (National University of Singapore)

BibTeX Citation

@inproceedings{wu_sigmod23,
        title = {{DeltaBoost: Gradient Boosting Decision Trees with Efficient Machine Unlearning}},
        author = {Wu, Zhaomin and Zhu, Junhui and Li, Qinbin and He, Bingsheng},
        series = {{SIGMOD} '23},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3589313},
        url = {https://dl.acm.org/doi/10.1145/3589313},
        year = {2023}
}

Incoming Citations (Sorted by Pagerank)

Showing 2 of 2 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
10,446	Eliminating Redundant Feature Tests in Decision Tree and Random Forest Inference on SQL Predicates	2026	SIGMOD	4.9793485e-05
11,621	Snapcase – Regain Control over Your Predictions with Low-Latency Machine Unlearning	2024	VLDB	4.9793485e-05

Outgoing Citations (Sorted by Pagerank)

Showing 2 of 2 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
4,037	HedgeCut: Maintaining Randomised Trees for Low-Latency Machine Unlearning	2021	SIGMOD	6.8390354e-05
6,321	"Amnesia" - A Selection of Machine Learning Models That Can Forget User Data Very Fast	2020	CIDR	5.814903e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	6,321	"Amnesia" - A Selection of Machine Learning Models That Can Forget User Data Very Fast	2020	CIDR
2	11,454	Datamap-Driven Tabular Coreset Selection for Classifier Training	2025	VLDB
3	8,000	Quantifying identifiability to choose and audit epsilon in differentially private deep learning	2021	VLDB
4	5,723	An Experimental Evaluation of Large Scale GBDT Systems	2019	VLDB
5	68	Privacy-Preserving Data Mining	2000	SIGMOD
6	4,037	HedgeCut: Maintaining Randomised Trees for Low-Latency Machine Unlearning	2021	SIGMOD
7	9,846	DimBoost: Boosting Gradient Boosting Decision Tree to Higher Dimensions	2018	SIGMOD
8	2,007	VF2Boost: Very Fast Vertical Federated Gradient Boosting for Cross-Enterprise Learning	2021	SIGMOD
9	1,277	Privacy Preserving Vertical Federated Learning for Tree-based Models	2020	VLDB
10	5,772	OpBoost: A Vertical Federated Tree Boosting Framework Based on Order-Preserving Desensitization	2023	VLDB