Sequential Processing in Neuroscience Inspired Models

In the past a variety of computational problems have been tackled with different neural network approaches. However, very little research has been done on a framework which connects neuroscience-oriented models with connectionist models and higher level symbolic processing. In this paper, we outline a framework which focuses on a hybrid integration of various neural and symbolic preference techniques in order to shed more light on how we may process higher level concepts, for instance for language processing based on concepts from neuroscience. It is a first hybrid framework which allows a link between various levels from neuroscience, connectionist Preference Moore machines and symbolic machines. Furthermore, we discuss a model of pulsed neural network for classification of natural language data.

@InProceedings{PW00, 
 	 author =  {Panchev, Christo and Wermter, Stefan},  
 	 title = {Sequential Processing in Neuroscience Inspired Models}, 
 	 booktitle = {Proceedings of Third International Workshop on Current Computational Architectures Integrating Neural Networks and Neuroscience},
 	 editors = {Stefan Wermter, Jim Austin, David Willshaw, Mark Elshaw},
 	 number = {},
 	 volume = {},
 	 pages = {84--88},
 	 year = {2000},
 	 month = {Jan},
 	 publisher = {},
 	 doi = {}, 
 }