Please use this identifier to cite or link to this item: http://hdl.handle.net/2445/98246
Title: Estudi sobre les xarxes neuronals artificials
Author: Navarro i López, Andrea
Director: Pujol Vila, Oriol
Keywords: Xarxes neuronals (Informàtica)
Tesis
Neurones
Aprenentatge automàtic
Algorismes computacionals
Neural networks (Computer science)
Theses
Neurons
Machine learning
Computer algorithms
Issue Date: 22-Jan-2016
Abstract: One of the main features of the living beings, in special the human, is the capacity of learning. And the fact is that the great dream of the artificial intelligence is to achieve the creation of machines as intelligent as any one of us. Even still being away from succeeding, many researchers in the matter think that the best way to achieve it is from algorithms of learning, that test to imitate how the human brain works. The artificial neural networks simulate the behaviour of the biological nervous system. It treats of a system of models of artificial neurons, interconnected, arranged in a network with the purpose to collaborate to attain their common aim: the learning. In spite of the differences of structure between the biological and the artificial system, in both cases we find a massive group of simple units of process, the neurons. And it is in their connections where the intelligence of the network is placed. We plan to tackle the backpropagation algorithm both in theoretical and practical approaches so we can build an object recognition system.
Note: Treballs Finals de Grau de Matemàtiques, Facultat de Matemàtiques, Universitat de Barcelona, Any: 2016, Director: Oriol Pujol Vila
URI: http://hdl.handle.net/2445/98246
Appears in Collections:Programari - Treballs de l'alumnat
Treballs Finals de Grau (TFG) - Matemàtiques

Files in This Item:
File Description SizeFormat 
memoria.pdfMemòria2.14 MBAdobe PDFView/Open
codi_font.zipCodi font35.12 kBzipView/Open


This item is licensed under a Creative Commons License Creative Commons