LEARNING ATTRACTORS IN AN ASYNCHRONOUS, STOCHASTIC ELECTRONIC NEURAL-NETWORK

LANN27 is an electronic device implementing in discrete electronics a fully connected (full feedback) network of 27 neurons and 351 plastic synapses with stochastic Hebbian learning. Both neurons and synapses a re dynamic elements, with two time constants-fast for neurons and slow for synapses. Learning, synaptic dynamics, is analogue and is driven in a Hebbian way by neural activities. Long-term memorization takes pl ace on a discrete set of synaptic efficacies and is effected in a stoc hastic manner. The intense feedback between the nonlinear neural eleme nts, via the learned synaptic structure, creates in an organic way a s et of attractors for the collective retrieval dynamics of the neural s ystem, akin to Hebbian learned reverberations. The resulting structure of the attractors is a record of the large-scale statistics in the un controlled, incoming flow of stimuli. As the statistics in the stimulu s flow changes significantly, the attractors slowly follow it and the network behaves as a palimpsest-old is gradually replaced by new. More over, the slow learning creates attractors which render the network a prototype extractor: entire clouds of stimuli, noisy versions of a pro totype, used in training, all retrieve the attractor corresponding to the prototype upon retrieval. Here we describe the process of studying the collective dynamics of the network, before, during and following learning, which is rendered complex by the richness of the possible st imulus streams and the large dimensionality of the space of states of the network. We propose sampling techniques and modes of representatio n for the outcome.