Fast and Communication-Efficient Algorithm for Distributed Support Vector Machine Training

Support Vector Machines (SVM) are widely used as supervised learning models to solve the classification problem in machine learning. Training SVMs for large datasets is an extremely challenging task due to excessive storage and computational requirements. To tackle so-called big data problems, one n...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE transactions on parallel and distributed systems Ročník 30; číslo 5; s. 1065 - 1076
Hlavní autoři: Dass, Jyotikrishna, Sarin, Vivek, Mahapatra, Rabi N.
Médium: Journal Article
Jazyk:angličtina
Vydáno: New York IEEE 01.05.2019
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:
Algorithms
Benchmark testing
classification algorithms
Communication
Convergence
Data management
distributed computing
Embedded systems
Internet of Things
iterative algorithms
Kernel
Machine learning
Matrix decomposition
message passing
multicore processing
Optimization
Parallel processing
parallel programming
quadratic programming
Support vector machines
Training
ISSN:1045-9219, 1558-2183
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Support Vector Machines (SVM) are widely used as supervised learning models to solve the classification problem in machine learning. Training SVMs for large datasets is an extremely challenging task due to excessive storage and computational requirements. To tackle so-called big data problems, one needs to design scalable distributed algorithms to parallelize the model training and to develop efficient implementations of these algorithms. In this paper, we propose a distributed algorithm for SVM training that is scalable and communication-efficient. The algorithm uses a compact representation of the kernel matrix, which is based on the QR decomposition of low-rank approximations, to reduce both computation and storage requirements for the training stage. This is accompanied by considerable reduction in communication required for a distributed implementation of the algorithm. Experiments on benchmark data sets with up to five million samples demonstrate negligible communication overhead and scalability on up to 64 cores. Execution times are vast improvements over other widely used packages. Furthermore, the proposed algorithm has linear time complexity with respect to the number of samples making it ideal for SVM training on decentralized environments such as smart embedded systems and edge-based internet of things, IoT.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1045-9219
1558-2183
DOI:10.1109/TPDS.2018.2879950