Year: 2010
Type: dissertation
Abstract: Since ages people have been wondering the inner workings of brain.\nThere is more and more information about processes involving particular\nbrain regions, yet synthesizing neural-networks in a large scale has always\nbeen a problem. Despite the fact that the popular desktop computer\nis becoming more powerful, simulating massively parallel computations\naways proved to be a challenge. Due to its genericness, in very specialized\nwork, such as neural networks, it is far better to design and use dedicated\narray of processors.\nSuch approach has been used in development of the Ubichip – a device\ndesigned specifically for such purposes. The design goal was to simplify the\nexecution elements to suit the needs of efficient neuron model algorithms\nemulation.\nAs with every development cycle of a complex tool there are many tasks\nthat may be carried out in parallel. This, of course, effects in allowing\nthe development team to shorten the cycle and provide the solution faster.\nThis however, requires a set of tools that will allow to prototype and verify\nwork progress against some set of basic rules.\nThe Ubichip was already supported by a toolkit named SpiNDeK,\nthat allowed to create networks and with the use of ModelSim simulate\nthe code. This solution for proof-checking the algorithms however proved\nto be cumbersome and due to many levels of indirection – slow. To increase\nefficiency when working with code, the programmer should not\nhave to wait endlessly watching the progress bars. To remedy this, improve\nefficiency and encourage more precise tuning and development of\nnew neuron-model algorithms the Ubichip Virtual Machine was born.\nAt first it was just meant to be a simple visualization tool, but as it\nturned out there was a missing link in the chain of tools available for the\nUbichip, which had to be filled.\nThus, in this work a virtual machine for the Ubichip has been developed,\nas well as a visualization tool that enables a convenient display of\nthe evolution of spiking neurons in a network.\ni
Author: Jan Maciej Budzisz
Language: English
Subfield: Hardware and Architecture
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2097620784