Bibliographic Metadata

Title
A physics approach to classical and quantum machine learning with applications in quantum experiment / Alexey A. Melnikov
AuthorMelnikov, Alexey A.
Thesis advisorBriegel, Hans J.
PublishedInnsbruck, July 2018
Description143 Seiten : Illustrationen, Diagramme
Institutional NoteUniversity of Innsbruck, Dissertation, 2018
Date of SubmissionJuly 2018
LanguageEnglish
Document typeDissertation (PhD)
Keywords (DE)Quantum machine learning / Quantum information / Machine learning / Reinforcement learning / Projective simulation / Quantum walks / Quantum experiments
Keywords (EN)Quantum machine learning / Quantum information / Machine learning / Reinforcement learning / Projective simulation / Quantum walks / Quantum experiments
URNurn:nbn:at:at-ubi:1-25030 Persistent Identifier (URN)
Restriction-Information
 The work is publicly available
Files
A physics approach to classical and quantum machine learning with applications in quantum experiment [8.31 mb]
Links
Reference
Classification
Abstract (German)

Nowadays machine learning plays an increasing role in everyday life. What can be the role of machine learning in quantum physics and, importantly, the role of quantum physics in machine learning? This thesis addresses these questions by connecting quantum physics and machine learning, specifically reinforcement learning, from two directions: reinforcement learning helps in solving quantum physics problems, and quantum effects improve the performance of reinforcement learning algorithms. Both directions are considered by using the projective simulation model.

Abstract (English)

Nowadays machine learning plays an increasing role in everyday life. What can be the role of machine learning in quantum physics and, importantly, the role of quantum physics in machine learning? This thesis addresses these questions by connecting quantum physics and machine learning, specifically reinforcement learning, from two directions: reinforcement learning helps in solving quantum physics problems, and quantum effects improve the performance of reinforcement learning algorithms. Both directions are considered by using the projective simulation model.

Stats
The PDF-Document has been downloaded 21 times.