An Extreme Learning Machine Based on Artificial Immune System

Autor: Min Yao, Shi-jian Li, Tian-qi Wu, Hui-yuan Tian
Jazyk: angličtina
Rok vydání: 2018
Předmět:
0209 industrial biotechnology
General Computer Science
Article Subject
Computer science
Generalization
General Mathematics
02 engineering and technology
Machine learning
computer.software_genre
lcsh:Computer applications to medicine. Medical informatics
Models
Biological
lcsh:RC321-571
Machine Learning
020901 industrial engineering & automation
Convergence (routing)
0202 electrical engineering
electronic engineering
information engineering
Animals
Humans
Layer (object-oriented design)
lcsh:Neurosciences. Biological psychiatry. Neuropsychiatry
Extreme learning machine
Artificial neural network
Artificial immune system
business.industry
General Neuroscience
General Medicine
Immune System
Benchmark (computing)
lcsh:R858-859.7
020201 artificial intelligence & image processing
Neural Networks
Computer
Hidden layer
Artificial intelligence
business
computer
Research Article
Zdroj: Computational Intelligence and Neuroscience, Vol 2018 (2018)
Computational Intelligence and Neuroscience
ISSN: 1687-5273
1687-5265
Popis: Extreme learning machine algorithm proposed in recent years has been widely used in many fields due to its fast training speed and good generalization performance. Unlike the traditional neural network, the ELM algorithm greatly improves the training speed by randomly generating the relevant parameters of the input layer and the hidden layer. However, due to the randomly generated parameters, some generated “bad” parameters may be introduced to bring negative effect on the final generalization ability. To overcome such drawback, this paper combines the artificial immune system (AIS) with ELM, namely, AIS-ELM. With the help of AIS’s global search and good convergence, the randomly generated parameters of ELM are optimized effectively and efficiently to achieve a better generalization performance. To evaluate the performance of AIS-ELM, this paper compares it with relevant algorithms on several benchmark datasets. The experimental results reveal that our proposed algorithm can always achieve superior performance.
Databáze: OpenAIRE
Externí odkaz: