Sparked-based Parallel Extreme Learning Machine
[1]Deng Wanyu,Li Li,Niu Huijuan.Sparked-based Parallel Extreme Learning Machine[J].Journal of Zhengzhou University (Engineering Science),2016,37(05):47-56.[doi:10.3969/ j.issn.1671 -6833.2016.05.010]
Copy
Journal of Zhengzhou University (Engineering Science)[ISSN
1671-6833/CN
41-1339/T] Volume:
37
Number of periods:
2016 05
Page number:
47-56
Column:
Public date:
2016-11-25
- Title:
- Sparked-based Parallel Extreme Learning Machine
- CLC:
- TP389.1
- Abstract:
- With the rapid expansion of data scale, the serial neural network structure based on a single machine is facing huge computing challenges, and it is difficult to meet the expansion requirements in real applications. Based on the extreme learning machine (extreme learning machine, ELM), based on the Spark parallel framework A parallel extremely fast neural network learning method is proposed, which is encapsulated with the unique RDD efficient data set management mechanism of the Spark platform, and the high-complexity matrix calculation in large-scale data is parallelized to achieve ELM accelerated solution. A set of Map and Reduce operations can complete the training of the algorithm. The experimental results on a large number of real data sets show that the parallel ELM algorithm based on Spark has achieved significant performance improvement compared with the serial ELM.
- References:
- -
- Similar References:
Memo
-
Last Update:
