Evolving Spiking Neural Parameters for Behavioral Sequences
| Autor: | Roger K. Moore, Thomas M. Poulsen |
|---|---|
| Rok vydání: | 2009 |
| Předmět: |
Physical neural network
Spiking neural network Artificial neural network business.industry Time delay neural network Computer science Machine learning computer.software_genre Recurrent neural network Genetic algorithm ComputingMethodologies_GENERAL Artificial intelligence business Stochastic neural network computer Nervous system network models |
| Zdroj: | Artificial Neural Networks – ICANN 2009 ISBN: 9783642042768 ICANN (2) |
| DOI: | 10.1007/978-3-642-04277-5_79 |
| Popis: | Sequential behavior has been the subject of numerous studies that involve agent simulations. In such research, investigators often develop and examine neural networks that attempt to produce a sequence of outputs. Results have provided important insights into neural network designs but they offer a limited understanding of the underlying neural mechanisms. It is therefore still unclear how relevant neural parameters can advantageously be employed to alter motor output throughout a sequence of behavior. Here we implement a biologically based spiking neural network for different sequential tasks and investigate some of the neural mechanisms involved. It is demonstrated how a genetic algorithm can be employed to successfully evolve a range of neural parameters for different sequential tasks. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|