Interpretable Constructive Algorithm for Random Weight Neural Networks

arXiv:2307.00185v2 Announce Type: replace
Abstract: In this paper, an interpretable construction method (IC) with geometric information is proposed to address a significant drawback of incremental random weight neural networks (IRWNNs), which is the difficulty in interpreting the black-box process of hidden parameter selection.The IC utilises geometric relationships to randomly assign hidden parameters, which improves interpretability. In addition, IC employs a node pooling strategy to select the nodes that will both facilitate network convergence. The article also demonstrates the general approximation …

abstract algorithm arxiv box construction cs.ai cs.lg hidden incremental information networks neural networks paper parameters process random relationships type