On the Emergence of Rules in Neural Networks

25 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

A Machine Learning Method for Extracting Symbolic Knowledge from Recurrent Neural Networks (2004, background): “Hanson and Negishi (2002) showed that abstract rules (i.e., grammars) can be extracted from sequences of arbitrary input symbols by inventing abstract representations that can also be generalized to new symbol sets or grammars.”
Self-refreshing memory in artificial neural networks: learning temporal sequences without catastrophic forgetting (2004, background): “…Altmann (2002) has recently proposed a very interesting way (based on pre-training on prior experience) in which the freezing process can be avoided (see also the recent work of Hanson and Negishi 2002, where it has been shown that transfer in a recurrent network does not require weight freezing).”