Rješavanje problema usmjeravanja vozila podržanim učenjem

undergraduate thesis

• Title (english): Solving vehicle routing problems with reinforcement learning
• Author: David Supančić
• Mentor: Mario Brčić (mentor)
• Committee member: Mario Brčić (predsjednik povjerenstva)
• Committee member: Mihael Kovač (član povjerenstva)
• Committee member: Kristijan Poje (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-11, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom završnom radu istražuje se primjena Pointer Network-a (PN) i podrža-
nog učenja u rješavanju problema usmjeravanja vozila. Pointer Network je neuronska
mreža koja prima ulaznu sekvencu 2D koordinata te generira izlaznu sekvencu najkra-
ćeg puta, dok RL omogućuje učenje optimalnih akcija tako što ažurira težine PN-a.
Kombinirajući ove dvije tehnike, postiže se poboljšanje u rješavanju problema usmje-
ravanja vozila. U radu se opisuje arhitektura Pointer Network-a, primjena podržanog
učenja i eksperimentalno se evaluira performansa na slučajno generiranim instancama
problema.

Abstract (english)

This bachelor’s thesis explores the application of Pointer Network (PN) and Re-
inforcement Learning to solving the vehicle routing problem. The thesis focuses on
the development of a solution using a Pointer Network, a neural network architecture
capable of taking an input sequence of 2D coordinates and generating an output sequ-
ence of minimal route length. RL techniques are employed to teach a PN optimal
actions by updating its weights. By combining Pointer Network and Reinforcement
Learning, the thesis aims to improve the efficiency and effectiveness of vehicle routing
solutions. The architecture and training process of the Pointer Network, as well as the
experimental evaluation of the proposed approach, are described and analyzed in the
thesis.

• Keywords: umjetna inteligencija; strojno učenje; duboko učenje; podržano učenje; Pointer Network; problem usmjeravanja vozila
• Keywords (english): artificial intelligence; machine learning; deep learning; reinforcement learning; Pointer Network; vehicle routing problem
• Language: croatian
• URN:NBN: urn:nbn:hr:168:426889
• 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: 2023-10-11 10:35:28