Neuroverkot ja AlphaGo Zero

Tutkielmassa perehdytään AlphaGo Zeron toimintaan ja oleellisimpiin siinä käytettyihin tekniikoihin. AlphaGo Zero on tekoäly go-lautapelille, jota on pidetty tekoälyille hankalana. Tärkeä syy AlphaGo Zeron menestykseen oli käyttää neuroverkkojen ja Monte Carlo -puuhaun yhdistelmää. Löydän tälle mene...

Täydet tiedot

Bibliografiset tiedot
Päätekijä: Merikivi, Mikko
Muut tekijät: Informaatioteknologian tiedekunta, Faculty of Information Technology, Informaatioteknologia, Information Technology, Jyväskylän yliopisto, University of Jyväskylä
Aineistotyyppi: Kandityö
Kieli:fin
Julkaistu: 2019
Aiheet:
Linkit: https://jyx.jyu.fi/handle/123456789/64688
_version_ 1826225803826823168
author Merikivi, Mikko
author2 Informaatioteknologian tiedekunta Faculty of Information Technology Informaatioteknologia Information Technology Jyväskylän yliopisto University of Jyväskylä
author_facet Merikivi, Mikko Informaatioteknologian tiedekunta Faculty of Information Technology Informaatioteknologia Information Technology Jyväskylän yliopisto University of Jyväskylä Merikivi, Mikko Informaatioteknologian tiedekunta Faculty of Information Technology Informaatioteknologia Information Technology Jyväskylän yliopisto University of Jyväskylä
author_sort Merikivi, Mikko
datasource_str_mv jyx
description Tutkielmassa perehdytään AlphaGo Zeron toimintaan ja oleellisimpiin siinä käytettyihin tekniikoihin. AlphaGo Zero on tekoäly go-lautapelille, jota on pidetty tekoälyille hankalana. Tärkeä syy AlphaGo Zeron menestykseen oli käyttää neuroverkkojen ja Monte Carlo -puuhaun yhdistelmää. Löydän tälle menetelmälle muitakin käyttökohteita. Kehitystä aikaisempaan AlphaGo-tekoälyyn nähden oli erityisesti pelaamaan oppiminen ilman ihmisten ohjausta, kahden neuroverkon yhdistäminen yhdeksi, joka antaa kaksi ulostuloa, sekä residuaalisten neuroverkkojen käyttäminen. This thesis explains how AlphaGo Zero and the most relevant of the techniques it utilizes work. AlphaGo Zero is an artificial intelligence for go, a board game that has been thought to be difficult for computers. An important reason for the success of AlphaGo Zero was how it combined neural networks and Monte Carlo tree search. I present other applications for this method as well. AlphaGo Zero had a number of improvements compared to its predecessor AlphaGo. The most important ones were that it learns without human guidance, that it combines two neural networks into one that has two outputs, and that it utilizes residual networks.
first_indexed 2024-09-11T08:50:36Z
format Kandityö
free_online_boolean 1
fullrecord [{"key": "dc.contributor.advisor", "value": "Kaijanaho, Antti-Juhani", "language": "", "element": "contributor", "qualifier": "advisor", "schema": "dc"}, {"key": "dc.contributor.author", "value": "Merikivi, Mikko", "language": "", "element": "contributor", "qualifier": "author", "schema": "dc"}, {"key": "dc.date.accessioned", "value": "2019-06-19T05:57:17Z", "language": null, "element": "date", "qualifier": "accessioned", "schema": "dc"}, {"key": "dc.date.available", "value": "2019-06-19T05:57:17Z", "language": null, "element": "date", "qualifier": "available", "schema": "dc"}, {"key": "dc.date.issued", "value": "2019", "language": "", "element": "date", "qualifier": "issued", "schema": "dc"}, {"key": "dc.identifier.uri", "value": "https://jyx.jyu.fi/handle/123456789/64688", "language": null, "element": "identifier", "qualifier": "uri", "schema": "dc"}, {"key": "dc.description.abstract", "value": "Tutkielmassa perehdyt\u00e4\u00e4n AlphaGo Zeron toimintaan ja oleellisimpiin siin\u00e4 k\u00e4ytettyihin tekniikoihin. AlphaGo Zero on teko\u00e4ly go-lautapelille, jota on pidetty teko\u00e4lyille hankalana. T\u00e4rke\u00e4 syy AlphaGo Zeron menestykseen oli k\u00e4ytt\u00e4\u00e4 neuroverkkojen ja Monte Carlo -puuhaun yhdistelm\u00e4\u00e4. L\u00f6yd\u00e4n t\u00e4lle menetelm\u00e4lle muitakin k\u00e4ytt\u00f6kohteita. Kehityst\u00e4 aikaisempaan AlphaGo-teko\u00e4lyyn n\u00e4hden oli erityisesti pelaamaan oppiminen ilman ihmisten ohjausta, kahden neuroverkon yhdist\u00e4minen yhdeksi, joka antaa kaksi ulostuloa, sek\u00e4 residuaalisten neuroverkkojen k\u00e4ytt\u00e4minen.", "language": "fi", "element": "description", "qualifier": "abstract", "schema": "dc"}, {"key": "dc.description.abstract", "value": "This thesis explains how AlphaGo Zero and the most relevant of the techniques it utilizes work. AlphaGo Zero is an artificial intelligence for go, a board game that has been thought to be difficult for computers. An important reason for the success of AlphaGo Zero was how it combined neural networks and Monte Carlo tree search. I present other applications for this method as well. AlphaGo Zero had a number of improvements compared to its predecessor AlphaGo. The most important ones were that it learns without human guidance, that it combines two neural networks into one that has two outputs, and that it utilizes residual networks.", "language": "en", "element": "description", "qualifier": "abstract", "schema": "dc"}, {"key": "dc.description.provenance", "value": "Submitted by Paivi Vuorio (paelvuor@jyu.fi) on 2019-06-19T05:57:17Z\nNo. of bitstreams: 0", "language": "en", "element": "description", "qualifier": "provenance", "schema": "dc"}, {"key": "dc.description.provenance", "value": "Made available in DSpace on 2019-06-19T05:57:17Z (GMT). No. of bitstreams: 0\n Previous issue date: 2019", "language": "en", "element": "description", "qualifier": "provenance", "schema": "dc"}, {"key": "dc.format.extent", "value": "33", "language": "", "element": "format", "qualifier": "extent", "schema": "dc"}, {"key": "dc.language.iso", "value": "fin", "language": null, "element": "language", "qualifier": "iso", "schema": "dc"}, {"key": "dc.rights", "value": "In Copyright", "language": "en", "element": "rights", "qualifier": null, "schema": "dc"}, {"key": "dc.subject.other", "value": "AlphaGo Zero", "language": "", "element": "subject", "qualifier": "other", "schema": "dc"}, {"key": "dc.subject.other", "value": "Monte Carlo -puuhaku", "language": "", "element": "subject", "qualifier": "other", "schema": "dc"}, {"key": "dc.subject.other", "value": "tietokone-go", "language": "", "element": "subject", "qualifier": "other", "schema": "dc"}, {"key": "dc.subject.other", "value": "vahvistusoppiminen", "language": "", "element": "subject", "qualifier": "other", "schema": "dc"}, {"key": "dc.subject.other", "value": "konvoluutioneuroverkot", "language": "", "element": "subject", "qualifier": "other", "schema": "dc"}, {"key": "dc.subject.other", "value": "residuaaliset neuroverkot", "language": "", "element": "subject", "qualifier": "other", "schema": "dc"}, {"key": "dc.title", "value": "Neuroverkot ja AlphaGo Zero", "language": "", "element": "title", "qualifier": null, "schema": "dc"}, {"key": "dc.type", "value": "bachelor thesis", "language": null, "element": "type", "qualifier": null, "schema": "dc"}, {"key": "dc.identifier.urn", "value": "URN:NBN:fi:jyu-201906193278", "language": "", "element": "identifier", "qualifier": "urn", "schema": "dc"}, {"key": "dc.type.ontasot", "value": "Bachelor's thesis", "language": "en", "element": "type", "qualifier": "ontasot", "schema": "dc"}, {"key": "dc.type.ontasot", "value": "Kandidaatinty\u00f6", "language": "fi", "element": "type", "qualifier": "ontasot", "schema": "dc"}, {"key": "dc.contributor.faculty", "value": "Informaatioteknologian tiedekunta", "language": "fi", "element": "contributor", "qualifier": "faculty", "schema": "dc"}, {"key": "dc.contributor.faculty", "value": "Faculty of Information Technology", "language": "en", "element": "contributor", "qualifier": "faculty", "schema": "dc"}, {"key": "dc.contributor.department", "value": "Informaatioteknologia", "language": "fi", "element": "contributor", "qualifier": "department", "schema": "dc"}, {"key": "dc.contributor.department", "value": "Information Technology", "language": "en", "element": "contributor", "qualifier": "department", "schema": "dc"}, {"key": "dc.contributor.organization", "value": "Jyv\u00e4skyl\u00e4n yliopisto", "language": "fi", "element": "contributor", "qualifier": "organization", "schema": "dc"}, {"key": "dc.contributor.organization", "value": "University of Jyv\u00e4skyl\u00e4", "language": "en", "element": "contributor", "qualifier": "organization", "schema": "dc"}, {"key": "dc.subject.discipline", "value": "Tietotekniikka", "language": "fi", "element": "subject", "qualifier": "discipline", "schema": "dc"}, {"key": "dc.subject.discipline", "value": "Mathematical Information Technology", "language": "en", "element": "subject", "qualifier": "discipline", "schema": "dc"}, {"key": "yvv.contractresearch.funding", "value": "0", "language": "", "element": "contractresearch", "qualifier": "funding", "schema": "yvv"}, {"key": "dc.type.coar", "value": "http://purl.org/coar/resource_type/c_7a1f", "language": null, "element": "type", "qualifier": "coar", "schema": "dc"}, {"key": "dc.rights.accesslevel", "value": "openAccess", "language": null, "element": "rights", "qualifier": "accesslevel", "schema": "dc"}, {"key": "dc.type.publication", "value": "bachelorThesis", "language": null, "element": "type", "qualifier": "publication", "schema": "dc"}, {"key": "dc.subject.oppiainekoodi", "value": "602", "language": "", "element": "subject", "qualifier": "oppiainekoodi", "schema": "dc"}, {"key": "dc.subject.yso", "value": "strategiapelit", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.subject.yso", "value": "koneoppiminen", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.subject.yso", "value": "neuroverkot", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.subject.yso", "value": "teko\u00e4ly", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.rights.url", "value": "https://rightsstatements.org/page/InC/1.0/", "language": null, "element": "rights", "qualifier": "url", "schema": "dc"}]
id jyx.123456789_64688
language fin
last_indexed 2025-02-18T10:56:30Z
main_date 2019-01-01T00:00:00Z
main_date_str 2019
online_boolean 1
online_urls_str_mv {"url":"https:\/\/jyx.jyu.fi\/bitstreams\/69b3db6b-cca6-4ec5-ba69-db15ea636393\/download","text":"URN:NBN:fi:jyu-201906193278.pdf","source":"jyx","mediaType":"application\/pdf"}
publishDate 2019
record_format qdc
source_str_mv jyx
spellingShingle Merikivi, Mikko Neuroverkot ja AlphaGo Zero AlphaGo Zero Monte Carlo -puuhaku tietokone-go vahvistusoppiminen konvoluutioneuroverkot residuaaliset neuroverkot Tietotekniikka Mathematical Information Technology 602 strategiapelit koneoppiminen neuroverkot tekoäly
title Neuroverkot ja AlphaGo Zero
title_full Neuroverkot ja AlphaGo Zero
title_fullStr Neuroverkot ja AlphaGo Zero Neuroverkot ja AlphaGo Zero
title_full_unstemmed Neuroverkot ja AlphaGo Zero Neuroverkot ja AlphaGo Zero
title_short Neuroverkot ja AlphaGo Zero
title_sort neuroverkot ja alphago zero
title_txtP Neuroverkot ja AlphaGo Zero
topic AlphaGo Zero Monte Carlo -puuhaku tietokone-go vahvistusoppiminen konvoluutioneuroverkot residuaaliset neuroverkot Tietotekniikka Mathematical Information Technology 602 strategiapelit koneoppiminen neuroverkot tekoäly
topic_facet 602 AlphaGo Zero Mathematical Information Technology Monte Carlo -puuhaku Tietotekniikka koneoppiminen konvoluutioneuroverkot neuroverkot residuaaliset neuroverkot strategiapelit tekoäly tietokone-go vahvistusoppiminen
url https://jyx.jyu.fi/handle/123456789/64688 http://www.urn.fi/URN:NBN:fi:jyu-201906193278
work_keys_str_mv AT merikivimikko neuroverkotjaalphagozero