Please use this identifier to cite or link to this item: https://doi.org/10.1609/aaai.v34i01.5422
Title: Privacy-preserving gradient boosting decision trees
Authors: Li, Q
Wu, Z
Wen, Z
He, B 
Keywords: cs.LG
cs.LG
cs.CR
stat.ML
Issue Date: 1-Jan-2020
Publisher: Association for the Advancement of Artificial Intelligence
Citation: Li, Q, Wu, Z, Wen, Z, He, B (2020-01-01). Privacy-preserving gradient boosting decision trees. AAAI 2020 - 34th AAAI Conference on Artificial Intelligence 34 (01) : 784-791. ScholarBank@NUS Repository. https://doi.org/10.1609/aaai.v34i01.5422
Abstract: The Gradient Boosting Decision Tree (GBDT) is a popular machine learning model for various tasks in recent years. In this paper, we study how to improve model accuracy of GBDT while preserving the strong guarantee of differential privacy. Sensitivity and privacy budget are two key design aspects for the effectiveness of differential private models. Existing solutions for GBDT with differential privacy suffer from the significant accuracy loss due to too loose sensitivity bounds and ineffective privacy budget allocations (especially across different trees in the GBDT model). Loose sensitivity bounds lead to more noise to obtain a fixed privacy level. Ineffective privacy budget allocations worsen the accuracy loss especially when the number of trees is large. Therefore, we propose a new GBDT training algorithm that achieves tighter sensitivity bounds and more effective noise allocations. Specifically, by investigating the property of gradient and the contribution of each tree in GBDTs, we propose to adaptively control the gradients of training data for each iteration and leaf node clipping in order to tighten the sensitivity bounds. Furthermore, we design a novel boosting framework to allocate the privacy budget between trees so that the accuracy loss can be further reduced. Our experiments show that our approach can achieve much better model accuracy than other baselines.
Source Title: AAAI 2020 - 34th AAAI Conference on Artificial Intelligence
URI: https://scholarbank.nus.edu.sg/handle/10635/215375
ISBN: 9781577358350
DOI: 10.1609/aaai.v34i01.5422
Appears in Collections:Staff Publications
Elements

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
1911.04209v4.pdf1.04 MBAdobe PDF

OPEN

PublishedView/Download

Page view(s)

83
checked on Feb 2, 2023

Download(s)

1
checked on Feb 2, 2023

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.