Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
412609 | Neurocomputing | 2012 | 10 Pages |
Abstract
The focus of this work is to investigate the generalisation capability of compact, solid-state synapses recently proposed by the authors. The synapses can be configured to yield a static or dynamic response. Empirical models of the Post Synaptic Response (PSP), derived from hardware simulations, are developed and embedded into the neural network toolbox in MATLAB. A network of these synapses was then used to solve benchmark problems using a well-established training algorithm where the performance metric was convergence time, accuracy and weight range; the Spike Response Model (SRM) was used to implement point neurons. Results are presented and compared with standard synaptic responses.
Keywords
Related Topics
Physical Sciences and Engineering
Computer Science
Artificial Intelligence
Authors
A. Ghani, L. McDaid, A. Belatreche, S. Hall, S. Huang, J. Marsland, T. Dowrick, A. Smith,