Fractal neural networks for short term memory

Synaptic connection matrix encodes limited information. It is well known that neural network memory with storage prescriptions based on Hebb´s rule will collapse as more patterns are stored. By requiring that old patterns be automatically forgotten and the memory recall only the most recently ones, a new short-term neural network memory based on Y. Baram´s fractal neural network is obtained. Comparison is drawn with Morris and Wong´s method and the experimental results are shown to be rather satisfactory and encouraging