Extreme Learning Machine With Affine Transformation Inputs in an Activation Function

The extreme learning machine (ELM) has attracted much attention over the past decade due to its fast learning speed and convincing generalization performance. However, there still remains a practical issue to be approached when applying the ELM: the randomly generated hidden node parameters without...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE transaction on neural networks and learning systems Ročník 30; číslo 7; s. 2093 - 2107
Hlavní autoři: Cao, Jiuwen, Zhang, Kai, Yong, Hongwei, Lai, Xiaoping, Chen, Badong, Lin, Zhiping
Médium: Journal Article
Jazyk:angličtina
Vydáno: United States IEEE 01.07.2019
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:
Affine transformation (AT) activation function
Affine transformations
Algorithms
Approximation algorithms
Artificial neural networks
Benchmark testing
Benchmarks
classification
Datasets
Distributed databases
Entropy
extreme learning machine (ELM)
Image classification
Learning algorithms
Learning systems
Maximum entropy
Object recognition
Parameters
regression
Robustness (mathematics)
Scaling
Training
Tuning
ISSN:2162-237X, 2162-2388, 2162-2388
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:The extreme learning machine (ELM) has attracted much attention over the past decade due to its fast learning speed and convincing generalization performance. However, there still remains a practical issue to be approached when applying the ELM: the randomly generated hidden node parameters without tuning can lead to the hidden node outputs being nonuniformly distributed, thus giving rise to poor generalization performance. To address this deficiency, a novel activation function with an affine transformation (AT) on its input is introduced into the ELM, which leads to an improved ELM algorithm that is referred to as an AT-ELM in this paper. The scaling and translation parameters of the AT activation function are computed based on the maximum entropy principle in such a way that the hidden layer outputs approximately obey a uniform distribution. Application of the AT-ELM algorithm in nonlinear function regression shows its robustness to the range scaling of the network inputs. Experiments on nonlinear function regression, real-world data set classification, and benchmark image recognition demonstrate better performance for the AT-ELM compared with the original ELM, the regularized ELM, and the kernel ELM. Recognition results on benchmark image data sets also reveal that the AT-ELM outperforms several other state-of-the-art algorithms in general.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:2162-237X
2162-2388
2162-2388
DOI:10.1109/TNNLS.2018.2877468