Please use this identifier to cite or link to this item:
|Title:||Neural computation : theory and practice behind brain||Authors:||Wang, Justin Chang Li.||Keywords:||DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems||Issue Date:||2010||Abstract:||Liquid State Machine is a relatively new system which is capable of recognising real-world temporal patterns on noisy continuous input streams. We will also investigate on its applicability for practical usage. By first looking at how the human brain model and biological neural network has brought about the development of artificial neural networks. We will also get to see how the spiking neuron model would have an advantage over conventional artificial neural network in classifying the real-world temporal patterns. In this project, we selected a practical input type for implementation which would be visual input, inspired by how human see. The challenges of such an implementation would be in creating a good encoding scheme to change the 2-dimensional input into spike train(s). A possible encoding scheme has been created in this project which has been successful in classifying the input. The implementation was created in the Matlab environment.||URI:||http://hdl.handle.net/10356/40145||Rights:||Nanyang Technological University||Fulltext Permission:||restricted||Fulltext Availability:||With Fulltext|
|Appears in Collections:||EEE Student Reports (FYP/IA/PA/PI)|
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.