Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Sign up for free
Menu
Search
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Pricing
Search
Sign in
Sign up for free
Graph Neural Networks のビジネス応用可能性 heterogeneo...
Search
Sponsored
·
Ship Features Fearlessly
Turn features on and off without deploys. Used by thousands of Ruby developers.
→
uchi_k
July 03, 2020
Programming
3.5k
1
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
Graph Neural Networks のビジネス応用可能性 heterogeneous graph と論文再現性について
uchi_k
July 03, 2020
More Decks by uchi_k
See All by uchi_k
ACL2020 Category Survey: Sentiment Analysis
uchi_k
2
3.5k
抽出的文書要約における hetero graph の応用 Heterogeneous Graph Neural Networks for Extractive Document Summarization
uchi_k
0
1.3k
前処理が単語埋め込みに与える影響 A Comprehensive Analysis of Preprocessing for Word Representation Learning in Affective Tasks
uchi_k
2
1.1k
ACL精神医療論文まとめ 8min LT
uchi_k
0
1.4k
【論文紹介】医用画像への転移学習の有効性について Transfusion: Understanding Transfer Learning for Medical Imaging
uchi_k
4
3.8k
Graph: A Survey of Graph Neural Networks, Embedding, Tasks and Applications
uchi_k
1
1.3k
Other Decks in Programming
See All in Programming
Agents on Rails - Rails at Scale 2026
irinanazarova
0
290
【高い買い物LT会】初任給で話題の国産フィジカルAIを買った話
akagami
PRO
0
180
App Intentsのビルドプロセスを支える技術
kntkymt
0
470
Augmenting AI with the Power of Jakarta EE
ivargrimstad
0
420
MVNOの申込からeSIM開通までをiOSアプリでつなぐ- 本人確認・MNP・通信事業者基盤をまたぐ実装
satotakeshi
0
510
大喜利で理解するLLM as a Judge / Understanding LLM-as-a-Judge through Ogiri
rockname
0
170
JRuby: Past, Present, and Future
headius
0
210
Workers Cache を知る
syumai
0
310
Go × SIMDで高速化するベクトル検索 ~ルーフラインモデルでSIMDが効く境界を探れ! ~
po3rin
1
4.9k
C#の現在地 進化の歴史と、AI時代の.NET Everywhere
neuecc
5
4.4k
更なる可用性を求めて、5年間運用したKotlinのアプリケーションをGoでリプレイスする話
ken_tunc
0
400
SREの越境 / SRE Collaboration
y0hgi
2
290
Featured
See All Featured
What's in a price? How to price your products and services
michaelherold
247
13k
The Success of Rails: Ensuring Growth for the Next 100 Years
eileencodes
47
8.3k
Understanding Cognitive Biases in Performance Measurement
bluesmoon
32
3k
Joys of Absence: A Defence of Solitary Play
codingconduct
1
530
Site-Speed That Sticks
csswizardry
13
1.5k
Redefining SEO in the New Era of Traffic Generation
szymonslowik
1
450
AI Search: Implications for SEO and How to Move Forward - #ShenzhenSEOConference
aleyda
1
1.4k
Rebuilding a faster, lazier Slack
samanthasiow
85
9.7k
Noah Learner - AI + Me: how we built a GSC Bulk Export data pipeline
techseoconnect
PRO
0
440
Skip the Path - Find Your Career Trail
mkilby
1
240
実際に使うSQLの書き方 徹底解説 / pgcon21j-tutorial
soudai
PRO
203
76k
Practical Orchestrator
shlominoach
192
12k
Transcript
(SBQI/FVSBM/FUXPSLTͷϏδωεԠ༻Մೳੑ IFUFSPHFOFPVTHSBQIͱจ࠶ݱੑʹ͍ͭͯ
ڮ ݎࢤ uchi_k @__uchi_k__ About me yuni, inc. ද nlpaper.challenge
ӡӦ Freelance Machine Learning ɹɹɹɹɹEngineer / Researcher former ະ౿16, ژେใӃ, FreakOut Machine Learning Engineer
About yuni େاۀ͔ΒελʔτΞοϓɺݚڀػؔͳͲ͔Βػցֶशؔ࿈ͷडୗ։ൃ Λߦ͖ͬͯ·ͨ͠ σʔλυϦϒϯͳͷͮ͘Γࣄۀͱͯ͠ɺΦϯϥΠϯʹΑΔύʔι φϥΠζ৸۩ͷ࡞͍ͬͯͨΓ͠·͢ ࠓɺIFUFSPHFOFPVTHSBQIͷΛ͠Α͏ͱࢥ͍ͬͯ·͢ ࡢۀͨ͠ɺγʔυظʹ͋ΔελʔτΞοϓʢࣾһ໊ʣͰ͢ ػցֶशºϚʔέςΟϯάྖҬͰαʔϏε։ൃΛ͍ͯ͠·͢
ࠓ͓͢Δ͜ͱ ϚʔέςΟϯάºػցֶश ࣮ੈքͷάϥϑͷੑ࣭ )FUFSPHFOFPVTHSBQI͔Βͷػցֶशͱ࣮Ԡ༻ จͷ࠶ݱੑʹ͍ͭͯ ·ͱΊͱ(//ք۾ʹؔͯ͠ࢥ͏͜ͱ
.BSLFUJOHʷ.BDIJOF-FBSOJOH ϚʔέςΟϯάྖҬʹଟ͘ͷϨΠϠʔ͕͋Γɺ ػցֶश͕ར༻͞Ε͍ͯΔͷଟ͋͘Δɻ https://www.smartinsights.com/managing-digital-marketing/marketing-innovation/15-applications-artificial-intelligence-marketing/ άϥϑσʔλ͕ଟ͘ɺকདྷతʹ(//͕༗ޮͳͷͰʁ ͬ͘͟ΓͰ͕͢ɺҰൠతͳϑϨʔϜϫʔΫʹԊͬͯλεΫΛհ͠·͢
3"$& • 3FBDIΠϯόϯυ ◦ 4/4ࢪࡦͳͲͷίϯςϯπϚʔέςΟϯάɺ4&0ɺϝσΟΞͳͲͰ ϢʔβʔΛࣗࣾʹ༠ಋ͢Δ ◦ ػցֶश͕ΘΕ͍ͯΔͷҎԼͷΑ͏ͳλεΫ ◦ ίϯςϯπϚʔέςΟϯάͷޮՌଌఆ
◦ ίϯςϯπ*OUFOUͷΩϡϨʔγϣϯʢϦίϝϯυʣ ◦ ӡ༻ܕࠂ • "DUؔ৺Λ࣋ͨͤΔ ◦ ༠ಋͨ͠જࡏސ٬ʹରͯ͠దͳࢪࡦɾಋઢΛઃܭ͢Δ ◦ ػցֶश͕ΘΕ͍ͯΔͷɺ ◦ ϦʔυείΞϦϯάʢજࡏސ٬ͷॱҐ͚ʣ ◦ ސ٬ߦಈɾίϯςϯπ࠷దԽʢ-10ͳͲʣ ◦ λʔήςΟϯάࠂ
3"$& • $POWFSUߪങʹಈ͔͢ ◦ ڵຯΛ࣋ͬͯ͘Εͨސ٬ʹߪೖ·ͰͷಋઢΛҾ͘ ◦ ػցֶश͕ΘΕ͍ͯΔͷɺ ◦ ಈతՁ֨ઃఆʢηʔϧͳͲͷࢪࡦؚΉʣ ◦
ಋઢઃܭʢνϟοτϘοτͳͲʣ ◦ ϦλʔήςΟϯάࠂ • &OHBHFސ٬ͱΑ͍ؔΛங͘ ◦ ܧଓతʹߪೖొΛͯ͠Β͏ͨΊɺྑ࣭ͳؔΛҡ࣋͢Δ ◦ ػցֶश͕ΘΕ͍ͯΔͷɺ ◦ ΤϯήʔδΦʔτϝʔγϣϯʢ୭ʹɺԿΛɺ͍ͭɺͷࣗಈԽʣ ◦ ΧελϚʔαϙʔτʢϘοτʣ
6TFS(FOFSBUFE$POUFOUT 6($ऩू 4/4֦ࢄ ݕࡧ ࠂࢪࡦ ՄࢹԽ 4/4Λىʹͨ͠ίϯςϯπϚʔέͷޮՌଌఆΛࢧԉ
ϝλύεΛ௨͕ͯۙؔ͠ఆٛͰ͖Δ NFUBQBUIWFD4DBMBCMF3FQSFTFOUBUJPO-FBSOJOHGPS)FUFSPHFOFPVT/FUXPSLT ϚʔέςΟϯάʹ͓͚ΔHSBQI ࣮ੈքͷάϥϑIFUFSPHFOFPVTͳͷ͕ଟ͍
)FUFSPHFOFPVT(SBQI"UUFOUJPO/FUXPSL 9JBP8BOH #FJKJOH6OJWFSTJUZPG1PTUTBOE5FMFDPNNVOJDBUJPOT FUBM888 DJUBUJPOT ݱ࣮ੈքͷIFUFSPHFOFPVTHSBQIʹର͠ɺϝλύεΛྡؔͱͨ͠ ֊BUUFOUJPOʹΑΔख๏Ͱ4P5"Λୡ )FUFSPHFOFPVTHSBQIʹ͓͚ΔϝλύεΛհͨ͠ྡؔΛఆٛ
)FUFSPHFOFPVT(SBQI"UUFOUJPO/FUXPSL 9JBP8BOH #FJKJOH6OJWFSTJUZPG1PTUTBOE5FMFDPNNVOJDBUJPOT FUBM888 DJUBUJPOT ϝλύεۙϊʔυϨϕϧͰͷBUUFOUJPOͱɺ ϝλύεϨϕϧͰͷBUUFOUJPOͷ֊తͳBUUFOUJPOΛఆٛ OPEFMFWFMBHHSFHBUJPOͰϝλύεۙΛॏཁʹԠͯ͡ू͢Δ ڞ௨ۭؒͷࣹӨI@J ΤοδͷॏཁFЇ@JK
ਖ਼نԽͨ͠ΤοδॏཁЋЇ@JK ϝλύεʹجͮ͘ಛຒΊࠐΈ[Ї@J
)FUFSPHFOFPVT(SBQI"UUFOUJPO/FUXPSL 9JBP8BOH #FJKJOH6OJWFSTJUZPG1PTUTBOE5FMFDPNNVOJDBUJPOT FUBM888 DJUBUJPOT λεΫʹΑͬͯϝλύεͷॏཁҟͳΔͨΊɺ ϝλύεͷॏཁΛTFNBOUJDBUUFOUJPOͱͯ͠ಋೖ TFNBOUJDMFWFMBHHSFHBUJPOͰϝλύεΛॏཁʹԠͯ͡ू͢Δ ୯.-1ͰOPEFFNCFEEJOHΛ ඇઢܗม͠ɺຒΊࠐΈͷྨࣅͱ͠
ͯॏཁΛܭࢉ ਖ਼نԽ ϝλύεॏཁͰBHHSFHBUJPO
)FUFSPHFOFPVT(SBQI"UUFOUJPO/FUXPSL 9JBP8BOH #FJKJOH6OJWFSTJUZPG1PTUTBOE5FMFDPNNVOJDBUJPOT FUBM888 DJUBUJPOT ϊʔυϨϕϧɺϝλύεϨϕϧͷॏཁ͕ݟΕΔͷͰɺͦͷ͋ͨΓͷղ ऍੑ͕ٻΊΒΕΔ߹ʹ͑ͦ͏ ϝλύεͷઃܭ͕ඞཁͳͷͰɺ͋ΔఔͷυϝΠϯࣝඞཁͦ͏ ࣮ݧ݁ՌΛݟΔݶΓϝλύεؒͷॏཁ͕͍͍ͩͨಉ͡ʹͳͬͯ͠· ͍ɺ݁ՌTFNBOUJDGVTJPO͕͋·Γҙຯͳ͍Α͏ʹݟ͍͑ͯΔͷͰɺ
ͬͱϝλύε͕όϦΤʔγϣϯ๛͔Ͱॏཁ͕େ͖͘ҟͳΔλεΫͰ ࢼͯ͠Έ͍ͨ
)FUFSPHFOFPVTHSBQIͷ࣮Ԡ༻ • )FUFSPHFOFPVT(SBQI/FVSBM/FUXPSLTGPS.BMJDJPVT "DDPVOU%FUFDUJPO ◦ ;JRJ-JV "OU'JOBODJBM4FSWJDFT(SPVQ FUBM $*,.
DJUBUJPOT ◦ "MJQBZͰѱҙͷ͋ΔΞΧϯτΛݕग़͢ΔͨΊʹɺσόΠεΞΫ ςΟϏςΟ͔ΒͳΔIFUFSPHFOFPVTHSBQI͔ΒͷػցֶशΛߦ͏ • .FUBQBUIHVJEFE)FUFSPHFOFPVT(SBQI/FVSBM/FUXPSLGPS *OUFOU3FDPNNFOEBUJPO ◦ 4IBPIVB'BO #FJKJOH6OJWFSTJUZPG1PTUTBOE 5FMFDPNNVOJDBUJPOT FUBM,%% DJUBUJPOT ◦ Ϣʔβʔ͕5BPCBPΛ։͍ͨͱ͖ʹɺաڈͷߦಈ͔ΒࣗಈతʹΫΤϦ ʢJOUFOUʣΛϦίϝϯυ͢Δ
%//ਪનք۾Ͱى͍ͬͯ͜Δ ৽نख๏ͷग़ݱස͕ߴ͍σʔληοτ͕ଟذʹΘͨΔ͜ͱ ͔Βɺ4P5"͕ඇৗʹ͍ʹ͘͘ͳ͍ͬͯΔ 3FD4ZT#FTU1BQFSͰɺۙͷ%//ܥਪનख๏Λ ࠶ݱ࣮͠ൺֱͨ݁͠Ռɺݹయతͳख๏ʹ͢Βউͯͳ͍ͷ͕΄ͱΜͲͩͬͨ ,%% 888 3FD4ZT 4*(*3ͷ%//ؔ࿈จຊͷࢼͰɺ ࠶ݱͰ͖ͨͷ͕ҎԼͷຊ͔͠ͳ͔ͬͨ
࠶ݱͰ͖ͯɺ΄ͱΜͲ͕ݹయతͳख๏ʹෛ͚ͯ͠·ͬͨ ਖ਼͍͠ϕʔεϥΠϯʢݹయత͕ͩڧྗͳख๏ʣΛਖ਼͘͠ௐ͠ɺ ൺֱ͠Α͏ɻ࣮ɺલॲཧɺϋΠύϥΛެ։͠Α͏ɻ
(//ͷʮਖ਼͍͠ϕʔεϥΠϯʯΛ࡞ΔऔΓΈ • "'BJS$PNQBSJTPOPG(SBQI/FVSBM/FUXPSLTGPS(SBQI $MBTTJpDBUJPO 'FEFSJDP&SSJDB FUBM *$-3 DJUBUJPOT ◦ %($//
%J⒎1PPM &$$ (*/ (SBQI4"(&ͷͭͷϞσϧΛɺͭ ͷϕϯνϚʔΫͰ࠶࣮ݧ ◦ (//ΛߏʹͱΒΘΕͳ͍ϕʔεϥΠϯͱൺֱ͢Δ͜ͱͰɺߏ ใ͕׆༻͞Ε͍ͯΔ͔Λݕূ • 0QFO(SBQI#FODINBSL%BUBTFUTGPS.BDIJOF-FBSOJOHPO (SBQIT 8FJIVB)V FUBMBS9JW DJUBUJPOT ◦ େنʢ࠷େԯϊʔυɺԯΤοδʣͰଟ༷ͳλεΫɾྖҬΛΧ όʔͨ͠ϕϯνϚʔΫσʔληοτΛఏҊ ◦ 044తʹӡ༻͞Ε͍ͯΔ
(//ͷʮਖ਼͍͠ϕʔεϥΠϯʯΛ࡞ΔऔΓΈ • #FODINBSLJOH(SBQI/FVSBM/FUXPSLT 7JKBZ1SBLBTI %XJWFEJ FUBMBS9JW DJUBUJPOT ◦ طଘख๏খنͳσʔλͰධՁ͞Ε͍ͯΔ͜ͱ͕ଟ͘ɺҰൠԽՄೳ ͳΞʔΩςΫνϟ͔ఆͮ͠Β͍
◦ HSBQISFHSFTTJPODMBTTJpDBUJPO OPEFDMBTTJpDBUJPO MJOL QSFEJDUJPOͷͭͷλεΫʹ͍ͭͯɺϕϯνϚʔΫͱͯ͠༻͢Δʹ ;͞Θ͍͠σʔληοτΛબఆ ◦ ($/T 8-(//Tͷطଘख๏Ͱ࠶ݱ࣮ݧΛͨ͠
#FODINBSLJOH(SBQI/FVSBM/FUXPSLT BS9JW %XJWFEJ 7JKBZ1SBLBTIBOE+PTIJ $IBJUBOZB,BOE-BVSFOU 5IPNBTBOE#FOHJP :PTIVBBOE#SFTTPO 9BWJFS DJUBUJPOT தنσʔλͰͷάϥϑճؼʗྨɺϊʔυϦϯΫ༧ଌͳͲͷλεΫͰɺ
طଘख๏ͷ࠶ݱ࣮Λൺֱͨ͠ɻ طଘख๏ͰΘΕ͍ͯΔσʔληοτ͕খن͗͢Δ "'BJS$PNQBSJTPOPG(SBQI/FVSBM/FUXPSLTGPS(SBQI$MBTTJpDBUJPO 'FEFSJDP&SSJDB FUBM *$-3
#FODINBSLJOH(SBQI/FVSBM/FUXPSLT BS9JW %XJWFEJ 7JKBZ1SBLBTIBOE+PTIJ $IBJUBOZB,BOE-BVSFOU 5IPNBTBOE#FOHJP :PTIVBBOE#SFTTPO 9BWJFS DJUBUJPOT தنσʔλͰͷάϥϑճؼʗྨɺϊʔυϦϯΫ༧ଌͳͲͷλεΫͰɺ
طଘख๏ͷ࠶ݱ࣮Λൺֱͨ͠ɻ (//T 8-(//Tͷ࣮ݧϑϩʔΛఆٛ
#FODINBSLJOH(SBQI/FVSBM/FUXPSLT BS9JW %XJWFEJ 7JKBZ1SBLBTIBOE+PTIJ $IBJUBOZB,BOE-BVSFOU 5IPNBTBOE#FOHJP :PTIVBBOE#SFTTPO 9BWJFS DJUBUJPOT தنσʔλͰͷάϥϑճؼʗྨɺϊʔυϦϯΫ༧ଌͳͲͷλεΫͰɺ
طଘख๏ͷ࠶ݱ࣮Λൺֱͨ͠ɻ άϥϑߏΛແࢹͯ͠.-1ͰϞσϧԽ͢ΔͱείΞ͕͍ͷͰɺάϥ ϑߏͷϞσϦϯά͕ඞਢͷλεΫͰ͋Γͦ͏ 8-(//Tɺ8-UFTUͱಉͷࣝผೳྗΛ࣋ͭͷͷɺ࠷ۙͷ 8-(//3JOH(//($/ΛείΞͰԼճΔ
(//จͷ࠶ݱੑ σʔληοτ͕খنͩͬͨΓҰൠԽͮ͠Β͔ͬͨΓͰɺ4P5" ͕͍ʹ͍͘ͱ͍͏%//ਪનͱಉ͕ͩ͡ɺ͜Ε͋Δఔ ͠ΐ͏͕ͳ͍໘͋Δ άϥϑΧʔωϧΛͬͨख๏ͳͲɺݹయతख๏ͱͷͪΌΜͱ͠ ͨൺֱݟͯΈ͍ͨɾɾɾ ίʔυ͕ެ։͞Ε͍ͯΔจ͕ଟ͘ɺடং͕อͨΕ͍ͯΔΑ͏ ʹײ͡·͢ʢࢲݟͰ͢ʣ
ύωϧσΟεΧογϣϯ IFUFSPHFOFPVTHSBQIͷख๏ʹଞʹͲΜͳͷ͕͋Δ ͷʁ ͦͦάϥϑߏͷຒΊࠐΈʹͲΜͳछྨ͕͋Δʁ ࠷ۙҰ൪໘ന͍ͱࢥͬͨจɾऔΓΈͳΜͰ͔͢ʁʢෳ ͋ͬͯେৎͰ͢ʂʣ