Uporaba algoritma podržanog učenja u problemima s većim prostorom stanja

undergraduate thesis

• Title (english): Application of reinforcement learning on large state-space problems
• Author: Stjepan Dugonjić
• Mentor: Marko Čupić (mentor)
• Committee member: Marko Čupić (predsjednik povjerenstva)
• Committee member: Darijan Marčetić (član povjerenstva)
• Committee member: Davor Kukolja (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2017-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Razvoj sustava koji sami uče jedan je od važnijih problema računarske znanosti. Poseban izazov predstavlja učenje rješavanja problema s velikim prostorom stanja. U ovom se radu razmatra implementacija Q-učenja, algoritma podržanog učenja uz aproksimaciju Q funkcije umjetnom neuronskom mrežom. Ispituje se sposobnost takvog sustava na primjeru savladavanja klasične igre Bomberman. Prikazani su i komentirani dobiveni rezultati, uz kratak opis funkcionalnosti razvijenog grafičkog sučelja.

Abstract (english)

Development of self-learning systems is one of the key aspects of computer science. Most challenging problems are ones where the state-space domain size is astronomically large. This thesis covers implementation of Q-learning, a reinforcement learning algorithm with non-linear Q function approximation using artificial neural networks. Capabilities of the algorithm are tested on the classic game Bomberman. Results are shown and commented, and short description of developed graphical user interface is presented.

• Keywords: Podržano učenje; neuronske mreže; Q-učenje; aproksimacija; velik prostor stanja
• Keywords (english): Reinforcement learning; Q-learning; neural network; approximation; large state-space
• Language: croatian
• URN:NBN: urn:nbn:hr:168:886267
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-13 20:03:45