Structural learning with forgetting

TL;DR: This article proposes to bring the various neuro-fuzzy models used for rule generation under a unified soft computing framework, and includes both rule extraction and rule refinement in the broader perspective of rule generation.

TL;DR: This paper introduces evolving fuzzy neural networks (EFuNNs) as a means for the implementation of the evolving connectionist systems (ECOS) paradigm that is aimed at building online, adaptive intelligent systems that have both their structure and functionality evolving in time.

TL;DR: In this paper, a saliency criterion based on connection sensitivity is proposed to identify structurally important connections in the network for a given task, and the retained connections are indeed relevant to the given task.

TL;DR: It is found that the evolutionary algorithm will converge incorrectly if the approximate model has false optima, so two strategies to control the evolution process are introduced and methods to eliminate false minima in neural network training are proposed.

TL;DR: A class of practical and nearly optimal schemes for adapting the size of a neural network by using second-derivative information to make a tradeoff between network complexity and training set error is derived.

TL;DR: The Cascade-Correlation architecture has several advantages over existing algorithms: it learns very quickly, the network determines its own size and topology, it retains the structures it has built even if the training set changes, and it requires no back-propagation of error signals through the connections of the network.

TL;DR: The approach taken by the methods described here is to train a network that is larger than necessary and then remove the parts that are not needed.