diplomsko delo
Julijana Djordjević (Author), Danijel Skočaj (Mentor)

Abstract

Diplomska naloga daje vpogled v vzporednice med tem, kako človeške in umetne nevronske mreže obdelujejo informacije. Preučimo ključne teme, da lahko razumemo odnos med temi vzporednicami. Začnemo z uvodom v nenehno razvijajoče se področje nevroznanosti. Sledi raziskovanje dela Davida Hubela in Torstena Wiesela, katerih poskusi so odkrili, kako se razvija vizualni sistem pri živih bitjih. Diplomska naloga nato obravnava kritična obdobja skupaj z njihovo povezavo z razvojem možganov in tveganji umetnih poskusov. S tem smo spoznali nekaj splošnih informacij o vizualnem zaznavanju ljudi. Sedaj sledi prehod na vizualno zaznavanje pri nevronskih mrežah. Začnemo z vpogledom v globoko učenje in njegovo povezavo z nevronskimi mrežami. Opišemo implementacijo CNN-jev in nato še vzpostavimo vzporednice s konceptom neokognitrona. Predstavljeno je tudi učenje s prenosom znanja. Za zaključek diplomska naloga obravnava relacijo med človeškimi in umetnimi nevronskimi mrežami.

Keywords

nevronske mreže;vizualni sistem;nevroznanost;računalništvo in informatika;visokošolski strokovni študij;diplomske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UL FRI - Faculty of Computer and Information Science
Publisher: [J. Djordjević]
UDC: 004.8:612.822(043.2)
COBISS: 75262723 Link will open in a new window
Views: 252
Downloads: 48
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Visual perception in humans and with artificial neural networks
Secondary abstract: This thesis gives an insight into the connection between how human and artificial neural networks process information. We investigate key topics to fully understand the relationship between these parallel lines. We start with an introduction to the ever-evolving field of neuroscience. This is followed by a dive into the work of David Hubel and Torsten Wiesel, whose experiments helped uncover how the visual system develops in living beings. The thesis then looks into critical periods along with their link to the development of the brain and the risks of laboratory experiments. At first we establish how humans process information, and then investigate how computers do the same. We begin with insights into deep learning and its connection to neural networks. We look at the implementation of CNNs and then further establish the parallel line connection through the concept of neocognitron. Transfer learning is also defined. Finally, the thesis links together the parallel line theories between living beings and computers.
Secondary keywords: neural networks;visual system;neuroscience;computer science;computer and information science;diploma thesis;Računalništvo;Umetna inteligenca;Nevroznanosti;Univerzitetna in visokošolska dela;
Type (COBISS): Bachelor thesis/paper
Study programme: 1000470
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages: 114 str.
ID: 13303750