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
ヅ大xはこみら 対バンLT <たつお>
Search
t4t5u0
December 24, 2019
Technology
140
1
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
ヅ大xはこみら 対バンLT <たつお>
LaTeXを触ってみた話
t4t5u0
December 24, 2019
More Decks by t4t5u0
See All by t4t5u0
FUN AI 親睦会LT
t4t5u0
0
35
FUN AI 新入生向け説明会資料
t4t5u0
0
140
童話_オオカミ少年__から学ぶエントロピー.pdf
t4t5u0
0
240
Other Decks in Technology
See All in Technology
Kiro Crewをはじめよう
kizawa2020
1
160
Rustuna: Rustによる高速なOptuna実装
pfn
PRO
0
200
夫婦で子育てするための育児標準化
codmoninc
PRO
0
430
ソフトウェア ファクトリー - 再現性のあるソフトウェアの工場づくり / Software factory
iktakahiro
6
5k
専門領域のAIをどう評価するか
bengo4com
2
270
DevelopersIO 2026 知識の地図を描く ─ オントロジーとは何か?
ishikawa_satoru
1
1.7k
無理せず続ける、子育てエンジニアの学び
codmoninc
PRO
0
960
Kiro Crewで「みのるん式」仕事術やってみた
minorun365
PRO
1
250
AI コーディングと人間の認知スループット / CTO Night & Day 2026 Morning Session / AI Coding and Human Cognitive Throughput
prog893
4
1.2k
⾮構造化データを価値に変える AI エージェント基盤とその組織
ivry_presentationmaterials
0
200
OptunaHub: ブラックボックス最適化のためのプラットフォーム
pfn
PRO
0
140
Gemini Intelligence 入門 ― AppFunctions と Gemini Nano
takuroabe
0
290
Featured
See All Featured
Evolving SEO for Evolving Search Engines
ryanjones
0
330
How to build an LLM SEO readiness audit: a practical framework
nmsamuel
2
950
[SF Ruby Conf 2025] Rails X
palkan
3
1.5k
The untapped power of vector embeddings
frankvandijk
2
1.9k
BBQ
matthewcrist
90
10k
Let's Do A Bunch of Simple Stuff to Make Websites Faster
chriscoyier
508
140k
The Spectacular Lies of Maps
axbom
PRO
1
1.1k
Writing Fast Ruby
sferik
630
63k
What Being in a Rock Band Can Teach Us About Real World SEO
427marketing
0
1.1k
Impact Scores and Hybrid Strategies: The future of link building
tamaranovitovic
0
460
Sam Torres - BigQuery for SEOs
techseoconnect
PRO
0
580
ラッコキーワード サービス紹介資料
rakko
1
5.4M
Transcript
L A TEX Λ৮ͬͯΈͨ ρେ x ະདྷେ ରόϯ LT ձ
͓ͨͭ ެཱͩͯ͜ະདྷେֶ December 24, 2019 1 / 11
࣍ 1 ࣍ 2 ࣗݾհ 3 L A TEX ͱ
4 L A TEX ͷ͍͍ͱ͜Ζ 5 L A TEX ͷͭΒ͍ͱ͜Ζ 6 ࣮ྫ 7 ऴΘΓʹ 2 / 11
ࣗݾհ ͓ͨͭ • ެཱͩͯ͜ະདྷେֶ ֶ෦ 1 • Twitter:@i4mwh4ti4m •
࠷ۙϋϚ͍ͬͯΔͷɿϩοΫ ▶ the shes gone ▶ ϚΧϩχ͑Μͽͭ ▶ ͱ͚ͨిٿ • ϝΠϯݴޠɿPython • ਂֶशɺ࣭มΛษڧத 3 / 11
L ATEX ͱ L A TEX ͱ • L A
TEX2ε ͕ݱߦͷόʔδϣϯ • L A TEX ͱ TEX Λجʹͨ͠จॻॲཧγεςϜͰ͋Δɽ • TEX ͱϑϦʔ൛γεςϜͰɺݴޠͰ͋Δɽ 4 / 11
L ATEX ͷ͍͍ͱ͜Ζ L A TEX ͷ͍͍ͱ͜Ζ • ͕͖ࣜΕ͍ʹॻ͚Δ •
ϕΫλʔը૾ׂ͕ͱ؆୯ʹॻ͚Δ • จॻ੶ͬΆ͍ࢿྉ͕࡞ΕΔ (ͦΕͦ͏) • ؒҧͬͯΔॴͪΌΜͱڭ͑ͯ͘ΕΔ 5 / 11
L ATEX ͷͭΒ͍ͱ͜Ζ L A TEX ͷͭΒ͍ͱ͜Ζ • ϚʔΫΞοϓʹͯ͠ίετ͕ॏ͍ •
ඇΤϯδχΞ͕ͱΓ͋͑ͣಈ͘ͷΛϊϋͱͯ͠ڞ༗ • ݹ͍ࢿྉ͕ଟ͍ • ڥߏங͕͍͠Β͍͠ 6 / 11
࣮ྫ ࣜ ͍͍ײ͡ʹࣜΛॻ͘͜ͱ͕Ͱ͖Δ i.e. ΧϧόοΫɾϥΠϒϥɾσΟόʔδΣϯε DKL(p||q) = − ∫ p(x)
ln q(x)dx − ( − ∫ p(x) ln p(x)dx ) = − ∫ p(x) ln { q(x) p(x) } dx = ∫ p(x) ln { p(x) q(x) } dx 7 / 11
࣮ྫ த \section{࣮ྫ} \begin{frame}{ࣜ} ͍͍ײ͡ʹࣜΛॻ͘͜ͱ͕Ͱ͖Δ\\ \textrm{i.e.} ΧϧόοΫɾϥΠϒϥɾσΟόʔδΣϯε \begin{align*} D_{\textrm{KL}}(p||q)&=-\int p(\textbf{x})\ln
q(\textbf{x})d\textbf{x} -\left(-\int p(\textbf{x})\ln p(\textbf{x})d\textbf{x}\right)\\ &=-\int p(\textbf{x})\ln\left\{\frac{q(\textbf{x})}{p(\textbf{x})}\right\} d\textbf{x}\\ &=\int p(\textbf{x})\ln\left\{\frac{p(\textbf{x})}{q(\textbf{x})}\right\} d\textbf{x} \end{align*} \end{frame} 8 / 11
࣮ྫ άϥϑ ࣜͱಉ༷ʹάϥϑΛॻ͘͜ͱͰ͖Δ x y O ς(x) tanh x ReLU
Figure: ओͳ׆ੑԽؔ 9 / 11
࣮ྫ த \begin{frame}{άϥϑ} ࣜͱಉ༷ʹάϥϑΛॻ͘͜ͱͰ͖Δ \begin{figure}[htbp]\centering \begin{tikzpicture} \draw[->, >=stealth, very thick]
(-5, 0) -- (5, 0) node[above]{$x$}; \draw[->, >=stealth, very thick] (0, -1.5) -- (0, 2.5) node[right]{$y$}; \draw (0,0)node[below right]{$O$}; \draw[samples=100,domain=-5:5,thick,red] plot(\x, {1/(1+exp(-(\x)))}) node[below]{$\varsigma(x)$}; \draw[samples=100,domain=-5:5,thick,green] plot(\x,{(exp(\x)-exp(-\x))/(exp(\x)+exp(-\x))})node[above]{$\tanh x$}; \draw[samples=100,domain=-5:2.5,thick,blue] plot(\x, {\x>0 ? \x : 0}) node[right]{$\textrm{ReLU}$}; \end{tikzpicture} \caption{ओͳ׆ੑԽؔ}\end{figure}\end{frame} 10 / 11
ऴΘΓʹ ऴΘΓʹ • Cloud LaTeX / Over leaf ͱ͔Λ͏ͱָ •
׳ΕΔͱ͍͍ײ͡ͷࢿྉ͕࡞ΕΔ • L A TEX ͕ॻ͚Δͱ͔͍͍ͬ͜ 11 / 11