Lihua Guo^*

Intelligent Automation & Soft Computing, Vol.26, No.3, pp. 421-427, 2020, DOI:10.32604/iasc.2020.013918

Abstract Compared with deep neural learning, the extreme learning machine (ELM) can be quickly converged without iteratively tuning hidden nodes. Inspired by this merit, an extreme learning machine with elastic net regularization (ELM-EN) is proposed in this paper. The elastic net is a regularization method that combines LASSO and ridge penalties. This regularization can keep a balance between system stability and solution's sparsity. Moreover, an excellent optimization method, i.e., accelerated proximal gradient, is used to find the minimum of the system optimization function. Various datasets from UCI repository and two facial expression image datasets are used to validate the efficiency of… More >

Jae-In Kim¹, Hyoseon Jang², Jeong-Hee Ahn³, Kilho Eom⁴, Sungsoo Na⁵

CMC-Computers, Materials & Continua, Vol.9, No.2, pp. 137-152, 2009, DOI:10.3970/cmc.2009.009.137

Abstract Protein dynamics is essential for gaining insight into biological functions of proteins. Although protein dynamics is well delineated by molecular model, the molecular model is computationally prohibited for simulating large protein structures. In this work, we provide the three-dimensional coarser-grained anisotropic model (CGAM), which is based on model reduction applicable to large protein structures. It is shown that CGAM achieves the fast computation on low-frequency modes, quantitatively comparable to original structural model such as elastic network model (ENM). This indicates that the CGAM by model reduction method enable us to understand the functional motion of large proteins with remarkable computational… More >

Yu Liu¹, Zhong Yang², Xiaoyong Yan³, Guangchi Liu⁴, Bo Hu^5,*

CMC-Computers, Materials & Continua, Vol.58, No.1, pp. 79-100, 2019, DOI:10.32604/cmc.2019.03626

Abstract Node location estimation is not only the promise of the wireless network for target recognition, monitoring, tracking and many other applications, but also one of the hot topics in wireless network research. In this paper, the localization algorithm for wireless network with unevenly distributed nodes is discussed, and a novel multi-hop localization algorithm based on Elastic Net is proposed. The proposed approach is formulated as a regression problem, which is solved by Elastic Net. Unlike other previous localization approaches, the proposed approach overcomes the shortcomings of traditional approaches assume that nodes are distributed in regular areas without holes or obstacles,… More >