Journal
IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS
Volume 33, Issue 11, Pages 3129-3140Publisher
IEEE COMPUTER SOC
DOI: 10.1109/TPDS.2021.3134336
Keywords
Clustering algorithms; Training; Mathematical models; Computational modeling; Libraries; Kernel; Support vector machines; Distributed machine learning; scalable algorithm; large-scale clustering; parallel regression
Funding
- National Natural Science Foundation of China [61972376, 62072431, 62032023]
- Science Foundation of Beijing [L182053]
Ask authors/readers for more resources
In this article, a novel data structure is proposed to capture the most important information among data samples, supporting a hierarchical clustering strategy. A parallel library that combines clustering and regression techniques is utilized to accelerate computation and improve accuracy.
As the data size in Machine Learning fields grows exponentially, it is inevitable to accelerate the computation by utilizing the ever-growing large number of available cores provided by high-performance computing hardware. However, existing parallel methods for clustering or regression often suffer from problems of low accuracy, slow convergence, and complex hyperparameter-tuning. Furthermore, the parallel efficiency is usually difficult to improve while striking a balance between preserving model properties and partitioning computing workloads on distributed systems. In this article, we propose a novel and simple data structure capturing the most important information among data samples. It has several advantageous properties supporting a hierarchical clustering strategy that contains well-defined metrics for determining optimal hierarchy, balanced partition for maintaining the clustering property, and efficient parallelization for accelerating computation phases. Then we combine the clustering with regression techniques as a parallel library and utilize a hybrid structure of data and model parallelism to make predictions. Experiments illustrate that our library obtains remarkable performance on convergence, accuracy, and scalability.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available