Experimental assessment of connectionest regular inference from positive and negative examples

Abstract

In this paper, the ability of recurrent neural networks (RNNs) for regular inference (RI) from positive and negative examples is investigated. As in some previous works [1, 2], RNNs were trained to learn the string classification task from samples of some target regular languages. In addition, an automaton extraction method [3] was applied to each one of the trained nets to obtain a description of the inference outcome. For comparison purposes, a symbolic RI method, the RPNI algorithm [4], was also run on the same data. Although the automaton extraction step improved the generalization performance of the nets, the inference quality using RNNs was not so good as the one shown by the RPNI algorithm.