Parsimonious Extreme Learning Machine Using Recursive Orthogonal Least Squares

• Published in: IEEE transaction on neural networks and learning systems Vol. 25; no. 10; pp. 1828 - 1841
• Main Authors: Wang, Ning, Er, Meng Joo, Han, Min
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.10.2014; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Context; Educational institutions; Elm; Error reduction; Extreme learning machine (ELM); Feedforward; Learning; Least squares method; Mathematical model; Matrix decomposition; Neural networks; parsimonious model selection; Recursive; recursive orthogonal least squares (ROLS); sequential partial orthogonalization (SPO); single hidden-layer feedforward network (SLFN); Training; Training data; Vectors; Weight reduction; Extreme learning machine (ELM); single hidden-layer feedforward network (SLFN); parsimonious model selection; sequential partial orthogonalization (SPO); recursive orthogonal least squares (ROLS)

Novel constructive and destructive parsimonious extreme learning machines (CP- and DP-ELM) are proposed in this paper. By virtue of the proposed ELMs, parsimonious structure and excellent generalization of multiinput-multioutput single hidden-layer feedforward networks (SLFNs) are obtained. The prop...