A Structure-Adaptive Hybrid RBF-BP Classifier with an Optimized Learning Strategy

Abstract

This paper presents a structure-adaptive hybrid RBF-BP (SAHRBF-BP) classifier with an optimized learning strategy. SAHRBF-BP is composed of a structure-adaptive RBF network and a BP network of cascade, where the number of RBF hidden nodes is adjusted adaptively according to the distribution of sample space, the adaptive RBF network is used for nonlinear kernel mapping and the BP network is used for nonlinear classification. The optimized learning strategy is as follows: firstly, a potential function is introduced into training sample space to adaptively determine the number of initial RBF hidden nodes and node parameters, and a form of heterogeneous samples repulsive force is designed to further optimize each generated RBF hidden node parameters, the optimized structure-adaptive RBF network is used for adaptively nonlinear mapping the sample space; then, according to the number of adaptively generated RBF hidden nodes, the number of subsequent BP input nodes can be determined, and the overall SAHRBF-BP classifier is built up; finally, different training sample sets are used to train the BP network parameters in SAHRBF-BP. Compared with other algorithms applied to different data sets, experiments show the superiority of SAHRBF-BP. Especially on most low dimensional and large number of data sets, the classification performance of SAHRBF-BP outperforms other training SLFNs algorithms.

Novel Concentric-Circle Technique Interrogates Implicit Category Learning

“…Though we made efforts to screen out low performers, it is well known that to date, online samples generally provide poorer data quality than in-lab samples Buchanan and Scofield (2018); Goodman and Wright (2022). Wen et al (2016). For example, every value of orientation (in isolation of frequency) of the inner circle, corresponds to a matching orientation on the outer circle; the same is true of frequency.…”

“…Our work was influenced by recent work in comparative psychology that pioneered a concentric ring category structure O'Donoghue, Broschard, Freeman, and Wasserman (2022). Earlier precedence exists within machine learning literature Wen, Xie, and Pei (2016). The finalized category structure proposed in our paper is similar to the one put forth by O'Donoghue and colleagues, though the methods by which the category sets are generated differ, as well as the nature of the investigation to which they are applied O' Donoghue et al (2022).…”

The pigeon as a machine: Complex category structures can be acquired by a simple associative model

“…Nature’s evolutionary process in creating associative learning may very well serve as an analogous model for computer scientists’ incorporating this powerful algorithm into their own artificial computing systems. 40 , 41 …”