Federated Doubly Stochastic Kernel Learning for Vertically Partitioned Data
Authors:
Open Access
- 20 August 2020
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 2483-2493
Abstract
In a lot of real-world data mining and machine learning applications, data are provided by multiple providers and each maintains private records of different feature sets about common entities. It is challenging to train these vertically partitioned data effectively and efficiently while keeping data privacy for traditional data mining and machine learning algorithms. In this paper, we focus on nonlinear learning with kernels,and propose a federated doubly stochastic kernel learning (FDSKL) algorithm for vertically partitioned data. Specifically, we use random features to approximate the kernel mapping function and use doubly stochastic gradients to update the solutions, which are all computed federatedly without the disclosure of data. Importantly, we prove that FDSKL has a sublinear convergence rate, and can guarantee the data security under the semi-honest assumption. Extensive experimental results on a variety of benchmark datasets show that FDSKL is significantly faster than state-of-the-art federated learning methods when dealing with kernels, while retaining the similar generalization performance.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- Privacy-Preserving Distributed Linear Regression on High-Dimensional DataProceedings on Privacy Enhancing Technologies, 2017
- Random Laplace Feature Maps for Semigroup Kernels on Histograms2014
- Parallel distributed computing using PythonAdvances in Water Resources, 2011
- LIBSVMACM Transactions on Intelligent Systems and Technology, 2011
- Distributed prediction from vertically partitioned dataJournal of Parallel and Distributed Computing, 2008
- Privacy-preservation for gradient descent methods2007
- Privacy-Preserving SVM Classification on Vertically Partitioned Data2006
- Privacy preserving regression modelling via distributed computation2004
- Privacy-preserving k -means clustering over vertically partitioned data2003
- Privacy preserving association rule mining in vertically partitioned data2002