Study on Deep Structure of Extreme Learning Machine (DS-ELM) for Datasets with Noise

Meng Meng Ma; Bo He

doi:10.4028/www.scientific.net/AMR.989-994.3679

Paper Titles

An Improved Image Stitching Algorithm
p.3663

Speech Airflow and Air-Pressure Signal Editing Platform
p.3667

Modeling Difficult-to-Measure Process Parameters Based on Intrinsic Mode Functions Frequency Spectral Features of Mechanical Vibration and Acoustical Signals
p.3671

Image Retrieval of Self-Adapt Distance Measure Based on SLLE
p.3675

Study on Deep Structure of Extreme Learning Machine (DS-ELM) for Datasets with Noise
p.3679

Design of 2_DOF Simulation Platform for the Testing of Airborne Laser Communication APT System
p.3683

The Method of Optimal Group Based on the Kernel Density Estimation and the Close Degree
p.3689

Research of Frequency-Hopping Communication for Arbitrary Array Based on Adaptive Beam-Forming
p.3693

Research on Fabric Image Acquisition Based on Compressed Sensing
p.3698

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 989-994Study on Deep Structure of Extreme Learning...

Study on Deep Structure of Extreme Learning Machine (DS-ELM) for Datasets with Noise

Abstract:

Extreme learning machine (ELM), a relatively novel machine learning algorithm for single hidden layer feed-forward neural networks (SLFNs), has been shown competitive performance in simple structure and superior training speed. To improve the effectiveness of ELM for dealing with noisy datasets, a deep structure of ELM, short for DS-ELM, is proposed in this paper. DS-ELM contains three level networks (actually contains three nets ): the first level network is trained by auto-associative neural network (AANN) aim to filter out noise as well as reduce dimension when necessary; the second level network is another AANN net aim to fix the input weights and bias of ELM; and the last level network is ELM. Experiments on four noisy datasets are carried out to examine the new proposed DS-ELM algorithm. And the results show that DS-ELM has higher performance than ELM when dealing with noisy data.