A plastic algorithm for building vector quantisers adaptively attains
a dynamic representation of observed data; an unsupervised version of
classical cross-validation rules the algorithm's stopping condition. C
ombining plasticity with empirical generalisation-based control yields
an adaptive methodology for ve. The paper analyses the method's conve
rgence properties and discusses the model's generalisation performance
. Experimental results on synthetic and real, complex testbeds support
the model's validity.