Simple evolving connectionist systems and experiments on isolated phoneme recognition

Evolving connectionist systems (ECoS) are systems that evolve their structure through online, adaptive learning from incoming data. This paradigm complements the paradigm of evolutionary computation based on population based search and optimisation of individual systems through generations of populations. The paper presents the theory and architecture of a simple evolving system called SECoS that evolves through one pass learning from incoming data. A case study of multi-modular SECoS systems evolved from a database of New Zealand English phonemes is used as an illustration of the method.