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
The beautiful world of evolutionary computation...
Search
S.Shota
October 09, 2017
Science
0
1.2k
The beautiful world of evolutionary computation made by probability and statistics
第二回Shot A Talkで使用したスライドになります
第二回Shot A Talk イベントページ:
https://shot-a-talk.connpass.com/event/60259/
S.Shota
October 09, 2017
Tweet
Share
More Decks by S.Shota
See All by S.Shota
[ICLR/ICML2019読み会] Adaptive Stochastic Natural Gradient Method for One-Shot Neural Architecture Search
satuma777
2
3.2k
論文紹介:Neural Architecture Search with Bayesian Optimisation and Optimal Transport [Kandasamy et al., NIPS 2018]
satuma777
0
1.1k
Path-Level Network Transformation for Efficient Architecture Search (ICML2018読み会)
satuma777
5
1.3k
論文紹介:Efficient Architecture Search by Network Transformation [Cai et al., AAAI-2018]
satuma777
0
1.1k
新ラボ生向けチュートリアル:文献調査(サーベイ)の仕方
satuma777
0
1.2k
編入試験への準備と編入後の生活 (Ver.2018)
satuma777
0
800
論文紹介:Gradient Boosted Feature Selection
satuma777
0
1.1k
論文紹介:Neural Architecture Search with Reinforcement Learning
satuma777
0
1.7k
論文紹介:Speeding up Automatic Hyperparameter Optimization of Deep Neural Networks by Extrapolation of Learning Curves
satuma777
0
1.2k
Other Decks in Science
See All in Science
データベース12: 正規化(2/2) - データ従属性に基づく正規化
trycycle
PRO
0
960
データベース08: 実体関連モデルとは?
trycycle
PRO
0
930
生成検索エンジン最適化に関する研究の紹介
ynakano
2
1.3k
Gemini Prompt Engineering: Practical Techniques for Tangible AI Outcomes
mfonobong
2
150
ウェブ・ソーシャルメディア論文読み会 第25回: Differences in misinformation sharing can lead to politically asymmetric sanctions (Nature, 2024)
hkefka385
0
130
学術講演会中央大学学員会府中支部
tagtag
0
300
Celebrate UTIG: Staff and Student Awards 2025
utig
0
120
データベース14: B+木 & ハッシュ索引
trycycle
PRO
0
430
IWASAKI Hideo
genomethica
0
130
Ignite の1年間の軌跡
ktombow
0
140
KH Coderチュートリアル(スライド版)
koichih
1
45k
Lean4による汎化誤差評価の形式化
milano0017
1
290
Featured
See All Featured
Mobile First: as difficult as doing things right
swwweet
223
9.9k
Imperfection Machines: The Place of Print at Facebook
scottboms
268
13k
Practical Tips for Bootstrapping Information Extraction Pipelines
honnibal
PRO
23
1.4k
For a Future-Friendly Web
brad_frost
179
9.9k
VelocityConf: Rendering Performance Case Studies
addyosmani
332
24k
Code Review Best Practice
trishagee
70
19k
Docker and Python
trallard
45
3.5k
Building an army of robots
kneath
306
46k
How GitHub (no longer) Works
holman
315
140k
How STYLIGHT went responsive
nonsquared
100
5.7k
Embracing the Ebb and Flow
colly
87
4.8k
We Have a Design System, Now What?
morganepeng
53
7.7k
Transcript
5IFCFBVUJGVMXPSMEPGFWPMVUJPOBSZDPNQVUBUJPO NBEFCZQSPCBCJMJUZBOETUBUJTUJDT 然桦ה鹌⻉鎘皾 _1ZUIPOח״知⽃侧⦼鎘皾_ 4IPUB4BJUP 俕谏 绤實 .BTUFSˏT$PVSTF4UVEFOU JO:PLPIBNB/BUJPOBM6OJWFSTJUZ *OUFMMJHFOU$PNQVUJOH-BC
1*4IJOJDIJ4IJSBLBXB -JLF .BDIJOF-FBSOJOH&WPMVUJPOBSZ$PNQVUBUJPO 8FC"QQMJDBUJPO1ZUIPO"JLB ,PCBZBTIJ
8IBUJT&WPMVUJPOBSZ$PNQVUBUJPO &$ • 1SPCBCJMJUZBOE4UBUJTUJDTBSFVTFEUP BQQSPYJNBUFCJPMPHJDBMFWPMVUJPO • %JTUJOHVJTICFUXFFOQSPCBCJMJUZEJTUSJCVUJPOT BDDPSEJOHUPUBSHFUGVODUJPO •
&$JTBQPXFSGVMGPS#MBDLCPY PQUJNJ[BUJPO ˏˏFWPMVUJPOBSZDPNQVUBUJPOJT BGBNJMZPGBMHPSJUINTGPSHMPCBMPQUJNJ[BUJPO JOTQJSFECZ CJPMPHJDBMFWPMVUJPO˒ IUUQTFOXJLJQFEJBPSHXJLJ&WPMVUJPOBSZ@DPNQVUBUJPO 4FF0DU
5IFXPSMEJTGJMMFEXJUICMBDLCPYFT #MBDL#PY1SPCMFN 4ZTUFN x y f ( x )
*OQVU 0VUQVU 0CTFSWBCMF /PU0CTFSWBCMFPS /PU%JGGFSFOUJBM 0QUJNJ[FUIFPVUQVUCZDPOUSPMMJOHUIFJOQVU x y
5IFXPSMEJTGJMMFEXJUICMBDLCPYFT "QQMJDBUJPOPG&$ • %FTJHOPG'SPOUOPTFJO#VMMFUUSBJO /TFSJFT • 3FEVDFESVOOJOHOPJTFBOE4QFFEVQ IUUQUSFOEZOJLLFJCQDPKQBSUJDMFDPMVNO 4FF0DU
$."&44UBUFPGUIFBSU3FBMWBMVFE&$ • *GJOQVUJT3FBMWBMVFEWFDUPS XFVTFT (BVTTJBO%JTUSJCVUJPO • $."&4JTUIFTUBUFPGUIFBSU&$CBTFEPO (BVTTJBO%JTUSJCVUJPO 4ZTUFN
x y f ( x ) 4BNQMJOH 6QEBUF%JTUSJCVUJPO (BVTTJBO
$."&44UBUFPGUIFBSU3FBMWBMVFE&$ 3FRVJSFEPG$PNQMFYNBUSJY WFDUPSPQFSBUJPOT IUUQTXXXMSJGS_IBOTFOHFDDP$."&4UVUPSJBMQEG 4FF0DU
&WPMUJFS1ZUIPO*NQMFNFOUBUJPOPG&$ • IUUQTHJUIVCDPNTBUVNBFWPMUJFS • HJWFNFBTUBS⭐ • &WPMUJFS QSPWJEFTBTJNQMFDPNNBOE"1* •
*NQMFNFOUUIFOFXFS&$CBTFEPO4UBUJTUJDBOE *OGPSNBUJPO(FPNFUSZ • 6TFGBTUOVNFSJDBMMJCSBSZˑ/VNQZ˒ • * QMBOUPTVQQPSU(16PQFSBUJPOJOUIFGVUVSF
&WPMUJFS1ZUIPO*NQMFNFOUBUJPOPG&$