Compact Extreme Learning Machines for biological systems. (17th September 2010)

• Record Type: Journal Article
• Authors: Li, Kang; Deng, Jing; He, Hai-Bo; Du, Da-Jun
• Is Part Of: International journal of computational biology and drug design. Volume 3:Number 2(2010)
• Page Start: 112
• Page End: 132
• Publication Date: 2010-09-17
• Subjects: ELM -- extreme learning machines -- fast recursive algorithms -- linear-in-the-parameter -- local regularisation -- radial basis function -- MAPK -- SOS response -- modelling -- biochemical data -- signal transduction pathway -- transcriptional network -- DNA damage -- E coli -- simulation -- computational biology; Computational biology -- Periodicals; Drugs -- Design -- Periodicals; 570.285
• Languages: English
• ISSNs: 1756-0756

Abstract

In biological system modelling using data-driven black-box methods, it is essential to effectively and efficiently produce a parsimonious model to represent the system behaviour. The Extreme Learning Machine (ELM) is a recent development in fast learning paradigms. However, the derived model is not necessarily sparse. In this paper, an improved ELM is investigated, aiming to obtain a more compact model without significantly increasing the overall computational complexity. This is achieved by associating each model term to a regularized parameter, thus insignificant ones are automatically unselected, leading to improved model sparsity. Experimental results on biochemical data confirm its effectiveness.