Abstract | ||
---|---|---|
An extension of an evolving spiking neural network (eSNN) is proposed that enables the method to process spatio-temporal information. In this extension, an additional layer is added to the network architecture that transforms a spatio-temporal input pattern into a single intermediate high-dimensional network state which in turn is mapped into a desired class label using a fast one-pass learning algorithm. The intermediate state is represented by a novel probabilistic reservoir computing approach in which a stochastic neural model introduces a non-deterministic component into a liquid state machine. A proof of concept is presented demonstrating an improved separation capability of the reservoir and consequently its suitability for an eSNN extension. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1007/978-3-642-17537-4_21 | ICONIP (1) |
Keywords | Field | DocType |
towards spatio-temporal pattern recognition,liquid state machine,stochastic neural model,esnn extension,spatio-temporal input pattern,intermediate state,novel probabilistic reservoir computing,network architecture,spiking neural network,spatio-temporal information,single intermediate high-dimensional network,proof of concept,reservoir computing | Pattern recognition,Computer science,Random neural network,Network architecture,Proof of concept,Liquid state machine,Time delay neural network,Artificial intelligence,Reservoir computing,Probabilistic logic,Spiking neural network,Machine learning | Conference |
Volume | ISSN | ISBN |
6443 | 0302-9743 | 3-642-17536-8 |
Citations | PageRank | References |
11 | 0.65 | 12 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Stefan Schliebs | 1 | 380 | 18.56 |
Nuttapod Nuntalid | 2 | 56 | 3.04 |
Nikola K Kasabov | 3 | 3645 | 290.73 |