Abstract
Models of temporal association in neural networks are generalised to provide mechanisms for decision making and loop control. Systems equipped with these capabilities can handle situations where a succession of states is not unambiguously defined and the system has to choose which out of several, in principle equivalent, paths it is going to follow. The choice is made on the basis of past experience of the network. No short-term synaptic plasticity is needed.
Original language | English |
---|---|
Pages (from-to) | 3123-3135 |
Number of pages | 13 |
Journal | Journal of Physics A: Mathematical and General |
Volume | 22 |
Issue number | 15 |
DOIs | |
State | Published - 7 Aug 1989 |
Externally published | Yes |