Title :
An episodic memory model using spiking neurons
Author_Institution :
Electrotech. Lab., Tsukuba, Japan
Abstract :
We describe a novel episodic memory model that meets some critical requirements for real-world robotic applications: (a) learn quickly and online, (b) recall patterns in their original order and with preserved timing information and (c) upon cuing, complete sequences from any position even in the presence of ambiguous transitions
Keywords :
bioelectric potentials; brain models; neural nets; robots; ambiguous transitions; critical requirements; episodic memory model; preserved timing information; real-world robotic applications; spiking neurons; Feedback loop; Feedforward systems; Firing; Hopfield neural networks; Neural networks; Neurons; Noise figure; Organisms; Robot sensing systems; Timing;
Conference_Titel :
Systems, Man, and Cybernetics, 2000 IEEE International Conference on
Conference_Location :
Nashville, TN
Print_ISBN :
0-7803-6583-6
DOI :
10.1109/ICSMC.2000.884969
