Primjena metoda dubokog učenja u razvoju investicijskih strategija

undergraduate thesis

• Title (english): Investment strategies development using deep learning methods
• Author: Mateo Stanić
• Mentor: Zvonko Kostanjčar (mentor)
• Committee member: Zvonko Kostanjčar (predsjednik povjerenstva)
• Committee member: Dražen Jurišić (član povjerenstva)
• Committee member: Ana Sović Kržić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2015-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

U okviru područja strojnog učenja, kao velikog dijela umjetne inteligencije, razvijaju se metode kojima je moguće implementirati sve složenije probleme. U ovom radu obrađena je teoretska pozadina na kojoj se temelji većina današnjih uspješnih implementacija. Obrađene su umjetne neuronske mreže, metode dubokog učenja kao potencijalnog rješenja prevelike dimenzionalnosti ulaznih podataka te Q – učenje kao metode podržanog učenja. Primjenjivost ovih metoda je ispitana u donošenju investicijskih strategija na burzi.

Abstract (english)

In the domain of machine learning, as a major section of artificial intelligence, numerous methods are being developed for solving and implementing complex problems. In this thesis, I have been writing about theoretical background of the machine learning problems which is the corner stone for almost every successful implementation today. Artificial neural networks, methods of deep learning as a solution to „curse of dimensionality“ and Q – learning as a method of reinforcement learning have been the main subjects.
Applicability of these methods has been tested on stock market investment decision making.

• Keywords: umjetna inteligencija; strojno učenje; portfelj; burza; umjetna neuronska mreža; duboko učenje; podržano učenje; Q - učenje; autoenkoder
• Keywords (english): artificial intelligence; machine learning; portfolio; stock market; artificial neural network; deep learning; reinforcement learning; Q - learning; autoencoder
• Language: croatian
• URN:NBN: urn:nbn:hr:168:546863
• Study programme: Title: Electrical Engineering and Information Technology; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka elektrotehnike i informacijske tehnologije (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka elektrotehnike i informacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2018-02-06 12:06:01