Year: 2018
Type: conference-paper
Abstract: Kernel methods have achieved tremendous success in the past two decades. In the current big data era, data collection has grown tremendously. However, existing kernel methods are not scalable enough both at the training and predicting steps. To address this challenge, in this paper, we first introduce a general sparse kernel learning formulation based on the random feature approximation, where the loss functions are possibly non-convex. Then we propose a new asynchronous parallel doubly stochastic algorithm for large scale sparse kernel learning (AsyDSSKL). To the best our knowledge, AsyDSSKL is the first algorithm with the techniques of asynchronous parallel computation and doubly stochastic optimization. We also provide a comprehensive convergence guarantee to AsyDSSKL. Importantly, the experimental results on various large-scale real-world datasets show that, our AsyDSSKL method has the significant superiority on the computational efficiency at the training and predicting steps over the existing kernel methods.
Authors: Bin Gu, Miao Xin, Zhouyuan Huo, Heng Huang
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
Awards: ABI Innovation: A New Automated Data Integration, Annotations, and Interaction Network Inference System for Analyzing Drosophila Gene Expression, III: Medium: Collaborative Research: Robust Large-Scale Electronic Medical Record Data Mining Framework to Conduct Risk Stratification for Personalized Intervention, G5520316217, BIGDATA: Collaborative Research: IA: Big Imaging-Omics Data Mining Framework for Precision Medicine, III: Small: Robust Large-Scale Data Mining for Knowledge Discovery in Depression Thought Records, SCH: EXP: Collaborative Research: Privacy-Preserving Framework for Publishing Electronic Healthcare Records, G9000211028, G9000243647
OpenAlex ID: W2789149658