Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
統計的因果推論勉強会第5回
Search
Sponsored
·
Your Podcast. Everywhere. Effortlessly.
Share. Educate. Inspire. Entertain. You do you. We'll handle the rest.
→
Hikaru Goto
October 29, 2016
Research
2.5k
0
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
統計的因果推論勉強会第5回
経営学系統計エンドユーザーのための統計的因果推論勉強会の第5回目です。これは公開用です。
Hikaru Goto
October 29, 2016
More Decks by Hikaru Goto
See All by Hikaru Goto
R実習 2016年9月25日
hikaru1122
1
2.7k
統計的因果推論勉強会 第4回
hikaru1122
0
1.5k
統計的因果推論勉強会 第3回
hikaru1122
0
2.1k
統計的因果推論勉強会 第2回
hikaru1122
0
2.2k
Other Decks in Research
See All in Research
National high-resolution cropland classification of Japan with agricultural census information and multi-temporal multi-modality datasets
satai
3
410
Geometric calculations on probability manifolds from reciprocal relations in Master equations
lwc2017
0
110
Sequences of Logits Reveal the Low Rank Structure of Language Models
sansantech
PRO
1
300
論文紹介 "ReSim: Reliable World Simulation for Autonomous Driving"
kogo
0
720
進学校の生徒にはア行の苗字が多いのか
ozekinote
0
520
Scalable dynamic origin-destination demand estimation enhanced by high-resolution satellite imagery data
satai
3
380
AIエージェント時代のLLM-jpモデルのあるべき姿
k141303
0
560
【ローカルAI LT大会】SSE: Stable Static Embedding ー速度低下を伴わず 静的埋め込みモデルの潜在能力を引き出す Dynamic Tanh手法の提案
rikkabotan7
0
110
GLIM とMegaParticles:正規分布近似の限界とタイトカップリング&パーティクルフィルタの進展 / GLIM and MegaParticles : Progress of the distribution representation in SLAM
koide3
0
680
SOTAのさらに先へ:厳しい推論制約下での高性能モデルのPost-Training
analokmaus
0
1.4k
「AIとWhyを深堀る」をAIと深堀る
iflection
0
550
多様なデータを許容し学習し続ける模倣学習 / Advanced Imitation Learning for VLA
prinlab
0
270
Featured
See All Featured
Evolution of real-time – Irina Nazarova, EuRuKo, 2024
irinanazarova
9
1.5k
Unsuck your backbone
ammeep
672
58k
Joys of Absence: A Defence of Solitary Play
codingconduct
1
430
Between Models and Reality
mayunak
4
390
Ecommerce SEO: The Keys for Success Now & Beyond - #SERPConf2024
aleyda
1
2.1k
Refactoring Trust on Your Teams (GOTO; Chicago 2020)
rmw
35
3.7k
GraphQLの誤解/rethinking-graphql
sonatard
75
12k
The browser strikes back
jonoalderson
0
1.5k
AI Search: Implications for SEO and How to Move Forward - #ShenzhenSEOConference
aleyda
1
1.3k
Improving Core Web Vitals using Speculation Rules API
sergeychernyshev
21
1.6k
The Cost Of JavaScript in 2023
addyosmani
55
10k
Odyssey Design
rkendrick25
PRO
2
750
Transcript
ܦӦֶܥ ౷ܭΤϯυϢʔβʔͷͨΊͷ ౷ܭతҼՌਪ ษڧձ ୈ5ճ 201610݄29 @hikaru1122 1
ษڧձͷϞοτʔ • ݚڀྗΛΞοϓ͠Α͏ɻ • ҼՌਪͷߟ͑ํɾ͍ํΛʹ͚ͭΑ͏ɻ • ֶతͳ͜ͱʹߦ͔ͳ͍ɻ 2
ຊͷൣғ • ٶຊɹୈ5ষ 81ʙ88ทʮόοΫυΞج४ʯ • ຊɹୈ4ষʮڞมྔબͱແࢹͰ͖ͳ͍ܽଌ 3
ຊͷత • ౷ܭతҼՌਪ͕ٻΊΒΕΔ݅ΛΔɻ • ҼՌޮՌΛٻΊΔͨΊͷڞมྔͷબͼํΛΔɻ • R ʹগ͠׳ΕΔɻ 4
ࠓͷ͓ଋ • ͕ݪҼมʢׂɾׂॲཧʣ • ͕݁Ռม • ͕ڞมྔʢަབྷҼࢠʣ 5
ͳͥڞมྔʹ͢Δͷ͔ʁ • ͔Β ͷҼՌޮՌΛΓ͍ͨͷʹɼଞͷཁૉ ͕มͳӨڹΛٴ΅͍ͯ͠Δ͔͠Εͳ͍ɻ • ͦΕΛίϯτϩʔϧͯ͠ɼ ͷ͚ؔͩΛ Γ͍ͨʂ 6
7
ࠓͷٶຊʮόοΫυΞج४ʯ • ճؼੳͷͱ͖ɼೖ͖͢આ໌มΛஅͰ͖ ΔΑ͏ʹͳΔɻ • όοΫυΞج४Λຬͨͨ͠มΛ͑ɼٖ૬ؔ ΛίϯτϩʔϧͰ͖Δɻຬ͍ͨͯ͠ͳ͍มΛ ͑ɼຊདྷͷҼՌޮՌ͕Θ͔Βͳ͘ͳΔɻ 8
όοΫυΞج४1 ࠓͷٶຊ͜Ε͚ͩͰेͰ͢ɻ • ΑΓ্ྲྀʹ͋Δɻதؒʹ͋Δͷμϝɻ • ͱ ͷ߹ྲྀͰͳ͍ɻ •
Λܦ༝͠ͳ͍ҹͰ ʹӨڹ͍ͯ͠Δɻ 1 ٶຊ 82,85ทͱྛɾࠇʢ2016ʣΛࢀߟʹ࡞ɻΑΓݫີͳఆٛٶຊΛࢀরͷ͜ͱɻ 9
ճؼੳΛ͢Δͱ͖ʹେͳ͜ͱ • ҼՌߏΛਤʹͯ͠ඳ͍ͯΈΔɻ • ੳʹ͍͍ͨݪҼมҎ֎ͷม͕όοΫυΞ ج४Λຬ͔ͨ͢ݕ౼͢Δɻ • ੳΛ࣮ߦʂ 10
όοΫυΞج४Λຬͨ͢ ͲΕʁ2 2 ྛɾࠇʢ2016ʣ͔ΒҾ༻ɻ͑ͱৄ͍͠ղઆͦͪΒΛࢀরɻ 11
12
13
ࠓͷຊʮڞมྔͷબʯ • ڞมྔʹείΞΛٻΊΔͱ͖ʹ͏આ໌ม • ݪҼมͱ݁ՌมʹӨڹΛ༩͑Δڞมྔͨ͘ ͞Μ͋ΔɻͲΕΛબ͍͍ͷʁ • ʮόοΫυΞج४ͳΜͯ͑ͳ͍͚Ͳͳʙʯ ʢ120ทʣ •
ͱݴ͑ɼࢲͨͪ͏ײతʹਤ4.1ͷҙຯΛ ཧղͰ͖Δʢ119ทʣɻ 14
15
ڞมྔͷબͼํ • ݁Ռมʹؔ࿈ʹࢥΘΕΔมɼதؒมͰ͋ Δ͜ͱʹҙ͠ͳ͕ΒɼͳΔ͘ଟ͘ೖ͢Δɻ • ਤ4.1ͷʢ̲ʣೖΕΔͱΑ͍ɼͱݴ͍ͬͯΔɻ • Γ͍ͨͷҼՌޮՌɻภճؼʹ͋·Γڵ ຯͳ͍ʢଟॏڞઢੑؾʹ͠ͳ͍ʣɻ •
͜Ε͕ٶຊͱͷҧ͍ɻ 16
ڧ͘ແࢹͰ͖ΔׂΓͯ݅ • ڞมྔௐʢڞมྔΛ৻ॏʹબͿʣͯ͠ɼ͜ͷ ͕݅ຬͨ͞Ε͍ͯͳ͍ͱμϝɻ • ͯ͢ͷڞมྔΛଌఆ͢Δ͜ͱͰ͖ͳ͍ɻ • ͔͠͠ʮڞมྔௐΛߦͬͨ΄͏͕ɼ୯७ͳ܈ؒ ൺֱΛߦ͏ΑΓ໌Β͔ʹҼՌޮՌʹ͍ۙਪఆΛ ༩͑Δʯ
• νΣοΫํ๏125ʙ126ทɻ 17
ڞมྔௐͷ࣮ࡍ • ͋·ΓͪΌΜͱߦΘΕ͍ͯͳ͍Α͏ͩʢ128 ทʣɻ • Ӝɾ࢞ʢ2015ʣ→ઌߦݚڀ͔ΒͷΈ • Տ߹Βʢ2016ʣ→ઌߦݚڀͱνΣοΫํ๏ʢ2ʣ 18
4.7ஶॻ͔Βͷϝοηʔδ • ώϧͷΨΠυϥΠϯҩֶܥͷจͰΑ͘ݟΔؾ ͕͢Δɻ • ࣜͳ͍ͷͰ҆৺ɻಡΜͰ͓͘ͱ౷ܭతҼՌਪ ͷཧղ͕ਂ·Δɻ 19
RʹΑΔ࣮श 20
ԿΛٻΊΔͷ͔ʁ • ࠓճATEΛٻΊΔɻ 21
ࢀߟจݙ • Pearl, J., Glymour, M. and Jewell, N. P.
(2016). Causal Inference in Statistics: A Primer. John Wiley & Sons. • ੴળथ, ࠓҪതٱ, தඌ༟೭, ᜊ౻૱, ా٢࣏. (2013). ಛఆอ݈ࢦಋͷ༧հೖࢪ ࡦͷޮՌʹؔ͢Δݚڀ: େنσʔλϕʔεΛ ༻ͨ͠είΞʹΑΔҼՌੳ. ްੜͷࢦ 22
• Ӝ, ࢞ګࢠ. (2015). େֶͷਐ ֶɾଔۀ͕ශࠔϦεΫʹ༩͑ΔޮՌ: εί ΞɾϚονϯά๏ʹΑΔߟ (ಛू ශࠔͷ৽
ͨͳࢹ). ౷ܭ, 66(5), 27-32. • ྛַɾࠇֶ(2016). ૬ؔͱҼՌͱؙͱҹ ͷͳ͠ɼؠσʔλαΠΤϯεɼvol.3ɼ 28ʙ48. 23
• ਸ(2010). ௐࠪ؍σʔλͷ౷ܭՊֶɹ ҼՌਪɾબόΠΞεɾσʔλ༥߹. ؠॻ ళ. • ਸ(2016). ౷ܭతҼՌޮՌͷجૅɼؠ σʔλαΠΤϯεɼvol.3ɼ62ʙ90.
• ٶխາ(2004). ౷ܭతҼՌਪʔճؼੳͷ ৽͍͠Έʔ. ேॻళ. 24