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
Hands-on-Machine-Learning Book Review
Search
Len Kim
March 01, 2019
Programming
40
1
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
Hands-on-Machine-Learning Book Review
책을 읽고 배울 수 있었던 점에 대해서 간략히 정리
Len Kim
March 01, 2019
More Decks by Len Kim
See All by Len Kim
오늘도 여러분의 API 는 안녕하신가요? - 김정규(인프콘2023)
lenkim
0
250
객체지향의 사실과오해 책 리뷰
lenkim
1
130
Other Decks in Programming
See All in Programming
Railsのためのパスキー実践入門 〜マルチテナントSaaSでの意思決定〜
utakaha
2
550
コードで差別化できない時代の ソフトウェアエンジニアリング / Software Engineering in an Era Where Code Doesn’t Differentiate
kishida
2
180
AIで開発が速くなった先で、開発チームは何を目指すか
shihochan
0
1k
JIS keyboard support in Omarchy
lef237
0
110
パズルゲームの作り方 / how to make puzzle games
kaityo256
PRO
2
290
iOS 27でニュースアプリはどう変わる!? 〜日経電子版の新機能対応と、開発事例から〜
lynnswap
7
13k
UnityでSystem.Net.WebSocketsなWebSocketサーバが動かないのでUnity Monoのコードを覗いてみた / about implementing websocket server with unity mono
drumath2237
1
610
IBM Bob Dojo #1 仕様駆動開発入門
oniak3ibm
PRO
0
390
Vue Fes Japan の “作る側”に関わってみて
naokihaba
0
160
AI主導のプログラミング学習アプリを作ってみた
econanringo
0
180
全ての永続トークンを、生まれる前に消し去りたい!
sylph01
2
880
Next.js: The final boss of Web Performance
guaca
0
260
Featured
See All Featured
For a Future-Friendly Web
brad_frost
184
10k
Code Review Best Practice
trishagee
74
20k
Claude Code どこまでも/ Claude Code Everywhere
nwiizo
68
59k
How to Create Impact in a Changing Tech Landscape [PerfNow 2023]
tammyeverts
56
3.5k
Raft: Consensus for Rubyists
vanstee
142
7.8k
Fashionably flexible responsive web design (full day workshop)
malarkey
409
67k
Digital Projects Gone Horribly Wrong (And the UX Pros Who Still Save the Day) - Dean Schuster
uxyall
1
3.4k
Performance Is Good for Brains [We Love Speed 2024]
tammyeverts
12
1.9k
Faster Mobile Websites
deanohume
310
32k
We Have a Design System, Now What?
morganepeng
55
8.4k
How GitHub (no longer) Works
holman
316
150k
Balancing Empowerment & Direction
lara
6
1.3k
Transcript
Hands-On Machine Learning ӣӏ
What is Machine Learning?
Goal • Machine Learning 의 정의를 내릴 수 있다. •
Machine Learning 프로젝트에 대한 프로세스를 이 해할 수 있다. (특히, CNN이 머신러닝의 처음이자 끝인줄 알고 있었던 사람들) • Machine Learning의 프로젝트 분류를 이해할 수 있다. • Machine Learning 프로젝트의 도전과제를 이해할 수 있다.
ӝദ ٣ੋ ѐߊ ߓನ ੌ߈ੋ ѐߊ ۽ࣁझ Ӓۧݶ, Machine Learning
۽ંח যڃ ۽ࣁझܳ о ө? ޙઁোҳ ӏ ࢿ (ࣛܖ࣌) ಣо ۠ட ML ѐߊ ۽ࣁझ ࣻࢎ೦ য়ର࠙ࢳ ݠन۞ ঌҊ્ܻ ള۲ ઁػ ؘఠ
분류
그렇다면 ML의 종류는 무슨 기준으로 나눌수 있을까?
지도, 비지도, 준지도, 강화 학습 ৡۄੋ णҗ ߓ ण ࢎ۹
ӝ߈ णҗ ݽ؛ ӝ߈ ण
ؘఠ ࣘী... ࢎۈ хة ೞী ള۲ೞח Ѫੋ Ӓۧ ঋ Ѫੋ
بण ࠺بण
ࢎۈ хة ೞী ള۲ೞח Ѫੋ Ӓۧ ঋ Ѫੋ ળبण ъചण
पदрਵ۽ ੋ णਸ ೞח ইצ ޙઁো ҳ (ࣛܖ࣌) ಣо ۠ட
য়ର࠙ ࢳ ݠन۞ ঌҊ્ܻ ള۲ ߓ ण ৡۄੋ ण ޙઁো ҳ (ࣛܖ࣌) ಣо ۠ட য়ର࠙ ࢳ ݠन۞ ঌҊ્ܻ ള۲ प೯ ߂ ঌҊܻ ્ ण ઁػ ؘఠ पदр ࢜۽ ؘఠ ઁػ ؘఠ
ױࣽೞѱ ঌҊ ח ؘఠ ನੋ৬ ࢜ ؘఠ ನੋܳ ࠺Үೞח Ѫੋ
ইפݶ ള۲ ؘఠࣇী җٜ ۢ ಁఢਸ ߊѼೞৈ ஏ ݽ؛ਸ ݅٘ח ࢎ۹ ӝ߈ ण ݽ؛ ӝ߈ ण
Machine Learning ਃ ب җઁ
?
나쁜 데이터
나쁜 데이터 충분하지 않은 양의 훈련 데이터 대표성 없는 훈련
데이터 낮은 품질의 데이터 관련 없는 특성 훈련 데이터 과대 적합 훈련 데이터 과소 적합
결론
아! 마지막 하나만 더! ҕबহۿ ݽ؛ҙஏೠѪਸрࣗചೠѪੑפрࣗചח࢜۽ࢠীੌ߈ ঋਸѪэࠛਃೠࣁࠗࢎ೦ਸઁѢೞחѪੑפӒ۞աযڃؘఠܳߡܻҊ যڃؘఠܳթӡೞӝਤ೧оਸ೧ঠפٜܳযࢶഋݽ؛ؘఠ оӔࠄਵ۽ࢶഋҊࢠҗࢶࢎѢܻחޖदೡࣻחۄҊо פ
֙ীߊೠਬݺೠ֤ޙীࢲؘ࠺٘ਘಌ%BWJE8PMQFSUTחؘఠীҙ೧ ৮߷ೞѱযڃоبೞঋਵݶೠݽ؛ਸܲݽ؛ࠁࢶഐೡӔѢоহਸࠁ णפܳҕबহ/P'SFF-VODI /'- ۿۄפযڃؘఠ ࣇীࢲחࢶഋݽ؛оੜٜযݏؘ݅ܲఠࣇীࢲחन҃ݎੜٜয ݏणפ҃ೞӝী؊ੜݏਸѢۄҊࠁೡࣻחݽ؛হणפ ۿܴਬېػਬੑפ যڃݽ؛୭ࢶੋഛपইחਬੌೠߑߨ ݽٚݽ؛ਸಣо೧ࠁחѪࡺੑפѪࠛоמೞӝٸޙীपীࢲחؘ ఠীҙ೧ఋೠоਸೞҊೠݽ؛ݻо݅ಣоפٜܳযрױೠ সীࢲחӏઁࣻળনೠࢶഋݽ؛ਸಣоೞҊ ࠂೠޙઁۄݶৈ۞о न҃ݎਸಣоפ