SONNET: a self-organizing neural network that classifies multiple patterns simultaneously
Albert Nigrin · 1990
The fundamentals are presented of a self-organizing neural network (SONNET) that can classify multiple distinct patterns simultaneously. The network consists of two fields,F(1)andF(2). Patterns are registered atF(1)and classified atF(2). The spatial patterns atF(1)continually evolve; therefore, learning must be done in realtime.F(2)is an on-center off-surround network that obeys winner-take-all dynamics. AtF(2), new classifications can form without degrading previous classifications; therefore, the learning is stable.F(2)is not a homogeneous field. Nodes learn different output characteristics so that different nodes can respond to different size patterns. Nonhomogeneous inhibitory connections form atF(2)so that nodes compete only with other nodes coding similar patterns. This allows multipleF(2)nodes (each representing a distinct pattern) to activate simultaneously