[B! æ©Ÿæ¢°å¦ç¿’][@] ZAORIKUã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

ZAORIKU id:ZAORIKU

æ©Ÿæ¢°å¦ç¿’ã¨@ã«é–¢ã™ã‚‹ZAORIKUã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (60)

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

${{author_name}}$
{{author_name}}{{created}}
{{ #comment }}{{ comment }}{{ /comment }}
- {{ label }}

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

åŒæ›²ç©ºé–“ã§ã®Machine Learningã®æœ€è¿‘ã®é€²å±• - ABEJA Tech Blog
ABEJAã§Reseacherã‚’ã—ã¦ã„ã‚‹ç™½å·ã§ã™ã€‚ ä»¥å‰ã€PoincarÃ© Embeddingsã¨ã„ã†åŒæ›²ç©ºé–“ã¸ã®åŸ‹ã‚è¾¼ã¿æ‰‹æ³•ã‚’ã”ç´¹ä»‹ã—ã¾ã—ãŸã€‚å½“æ™‚ã€æœ¨æ§‹é€ ãƒ‡ãƒ¼ã‚¿ã‚’5æ¬¡å…ƒã®ç©ºé–“ã«ç²¾åº¦è‰¯ãåŸ‹ã‚è¾¼ã‚ã‚‹ã¨ã„ã†ã“ã¨ã§è©±é¡Œã«ãªã£ãŸã®ã§ã™ãŒã€ãã®å¾ŒåŒæ›²ç©ºé–“ã§ã®æ©Ÿæ¢°å¦ç¿’æ‰‹æ³•ãŒå¤šæ•°ç ”ç©¶ãƒ»ææ¡ˆã•ã‚Œã€åŒæ›²ç©ºé–“ã§ã®æ©Ÿæ¢°å¦ç¿’ã«ã¤ã„ã¦ã®ç†è§£ã‚’ãƒãƒ¼ã‚¸ãƒ§ãƒ³ã‚¢ãƒƒãƒ—ã™ã‚‹å¿…è¦ãŒã‚ã‚‹ãªã¨æ„Ÿã˜ãŸã®ã§ã€æœ€è¿‘ã®ç ”ç©¶ã®é€²å±•ã‚’ä¸å¿ƒã«ç†è«–èƒŒæ™¯å«ã‚ã¦ã”ç´¹ä»‹ã—ãŸã„ã¨æ€ã„ã¾ã™ã€‚ tech-blog.abeja.asia Tl;dr æœ¬è¨˜äº‹ã§ä¼ãˆãŸã„ã®ã¯ã€è«–æ–‡ã®å„è«–ã¨ã„ã†ã‚ˆã‚Šã€å„ç¨®è«–æ–‡ã§å…±é€š/ç‹¬è‡ªã«ä¸»å¼µã•ã‚Œã¦ã„ã‚‹ä¸‹è¨˜ã®ã‚ˆã†ãªå†…å®¹ã§ã™ã€‚ æœ¨ãªã‚‰2æ¬¡å…ƒã§ååˆ† åŒæ›²ç©ºé–“ã§ã¯æŒ‡æ•°å†™åƒ/å¯¾æ•°å†™åƒãŒæ˜Žç¤ºçš„ã«è¨ˆç®—ã•ã‚Œç©ºé–“å…¨ä½“ã«æ‹¡å¼µã•ã‚Œã‚‹ã®ã§å–æ‰±ãŒå®¹æ˜“ Gyrovector space: åŒæ›²ç©ºé–“ã«ãŠã‘ã‚‹ç·šå½¢ä»£æ•°ã®ã‚ˆã†ãªä»£æ•°æ§‹é€ Riemannå¹¾ä½•ã¨Gyrove
ZAORIKU 2019/06/13
æ©Ÿæ¢°å¦ç¿’

@

è«–æ–‡
ãƒªãƒ³ã‚¯
XGBoostã®è«–æ–‡ã‚’èªã‚“ã ã®ã§GBDTã«ã¤ã„ã¦ã¾ã¨ã‚ãŸ - Dropout
ã¯ã˜ã‚ã« XGBoostè«–æ–‡ ç›®çš„é–¢æ•°ã®è¨å®š å‹¾é…ãƒ–ãƒ¼ã‚¹ãƒ†ã‚£ãƒ³ã‚° ã¾ã¨ã‚ å‚è€ƒæ–‡çŒ® ã¯ã˜ã‚ã« ä»Šæ›´ã§ã™ãŒã€XGboostã®è«–æ–‡ã‚’èªã‚“ã ã®ã§ã€2ç« GBDTéƒ¨åˆ†ã®ã¾ã¨ã‚è¨˜äº‹ã‚’æ›¸ã“ã†ã¨æ€ã„ã¾ã™ã€‚*1 ã“ã®è¨˜äº‹ã‚’æ›¸ãã«ã‚ãŸã£ã¦ã€ã§ãã‚‹ã ã‘æ•°å¼ã®è§£é‡ˆã‚’æ›¸ãã‚ˆã†ã«å¿ƒãŒã‘ã¾ã—ãŸã€‚æ•°å¼ã®æ„å‘³ã‚’ã²ã¨ã¤ã²ã¨ã¤è¿½ã£ã¦ã„ãã“ã¨ã¯ã€å®Ÿéš›ã«XGBoostï¼ˆã¾ãŸã¯LightGBMã‚„CatBoostãªã©ã®GBDTå®Ÿè£…ï¼‰ã‚’ä½¿ã†éš›ã«ã‚‚å½¹ç«‹ã¤ã¨è€ƒãˆã¦ã„ã¾ã™ã€‚ãŸã¨ãˆã°ãƒã‚¤ãƒ‘ãƒ¼ãƒ‘ãƒ©ãƒ¡ãƒ¼ã‚¿ãŒã©ã“ã«åŠ¹ã„ã¦ã„ã‚‹ã‹ã‚’ç†è§£ã™ã‚‹ã“ã¨ã§ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°ã‚’åŠ¹çŽ‡åŒ–ã—ãŸã‚Šã€ãƒ¢ãƒ‡ãƒ«ã‚’ç†è§£ã™ã‚‹ã“ã¨ã§ã‚ˆã‚Šãƒ¢ãƒ‡ãƒ«ã«åˆã£ãŸç‰¹å¾´é‡ã®ã‚¨ãƒ³ã‚¸ãƒ‹ã‚¢ãƒªãƒ³ã‚°ãŒã§ãã‚‹ã®ã§ã¯ãªã„ã‹ã¨æ€ã„ã¾ã™ã€‚ ã¾ãŸã€ã“ã®è¨˜äº‹ã«é™ã‚Šã¾ã›ã‚“ãŒã€è¨˜è¿°ã«é–“é•ã„ã‚„ä¸ååˆ†ãªç‚¹ãªã©ã‚ã‚Œã°ã”æŒ‡æ‘˜é ‚ã‘ã‚Œã°å¬‰ã—ã„ã§ã™ã€‚ XGBoostè«–æ–‡ ç›®çš„é–¢æ•°ã®è¨å®š ä¸€èˆ¬çš„ãªçŠ¶æ³ã¨ã—ã¦ã€ã‚µãƒ³ãƒ—ãƒ«ã‚µã‚¤ã‚ºãŒã§ç‰¹å¾´é‡ã®æ•°ãŒã®ãƒ‡ãƒ¼ã‚¿ã«
ZAORIKU 2019/05/08
@

æ©Ÿæ¢°å¦ç¿’
ãƒªãƒ³ã‚¯
ä¸å‡è¡¡ãƒ‡ãƒ¼ã‚¿ã¸ã®æ±ºå®šæ‰“ã¨ãªã‚‹ã‹ï¼ï¼Ÿã€ŒAffinity lossã€ã®è«–æ–‡ã‚’èªã‚€ã€å®Ÿè£…ã™ã‚‹ - Qiita
é™°æ€§ã®ãƒ‡ãƒ¼ã‚¿ã¯1ä¸‡ä»¶è¿‘ãé›†ã‚ã‚‹ã“ã¨ãŒã§ãã¾ã—ãŸãŒã€é™½æ€§ã®ãƒ‡ãƒ¼ã‚¿ã¯100ä»¶ã—ã‹é›†ã‚ã‚‰ã‚Œã¾ã›ã‚“ã§ã—ãŸã€‚ã“ã®ã‚ˆã†ã«ã€é™½æ€§ãƒ»é™°æ€§ã¨ã„ã†ã‚¯ãƒ©ã‚¹é–“ã§ãƒ‡ãƒ¼ã‚¿æ•°ã«æ˜Žã‚‰ã‹ãªåã‚ŠãŒã‚ã‚‹ã‚±ãƒ¼ã‚¹ã‚’ä¸å‡è¡¡ãƒ‡ãƒ¼ã‚¿ï¼ˆinbalanced dataï¼‰ã¨ã„ã„ã¾ã™ã€‚ ä¸å‡è¡¡ãƒ‡ãƒ¼ã‚¿ã®å•é¡Œç‚¹ è©•ä¾¡ä¸Šã®å•é¡Œ ã“ã‚Œã¯æ•™ç§‘æ›¸çš„ãªå•é¡Œã§ã€ã€Œç²¾åº¦ï¼ˆAccuracyï¼‰ãŒæ©Ÿèƒ½ã—ãªã„ã€ã¨ã„ã†ç‚¹ã§ã™ã€‚ä¾‹ãˆã°å…ˆç¨‹ã®ä¾‹ã§é™°æ€§ãŒãƒ‡ãƒ¼ã‚¿å…¨ä½“ã®99%ãªã‚‰ã€ã‚ã‚Šã¨ã‚ã‚‰ã‚†ã‚‹ãƒ‡ãƒ¼ã‚¿ã«å¯¾ã—ã¦é™°æ€§ã¨åˆ¤å®šï¼ˆ$y=0$ï¼‰ã™ã‚‹åˆ†é¡žå™¨ã‚’ä½œã‚Œã°ã€ãã®æ™‚ç‚¹ã§ç²¾åº¦99%ã‚’é”æˆã—ã¦ã—ã¾ã„ã¾ã™ã€‚ ã“ã‚“ãªã‚¤ãƒ¡ãƒ¼ã‚¸ã§ã™ã€‚æ©Ÿæ¢°å¦ç¿’ãªã‚“ã¦ã„ã‚‰ãªã„ã§ã™ãã€‚ç¬‘ã£ã¡ã‚ƒã†ã‹ã‚‚ã—ã‚Œã¾ã›ã‚“ãŒã€å®Ÿã¯æ©Ÿæ¢°å¦ç¿’ã‚’ä½¿ã£ã¦ã‚‚ã€Œreturn 0ã€ã¿ãŸã„ãªçŠ¶æ…‹ã«æœ€é©åŒ–ãŒå‘ã‹ã†ã“ã¨ãŒã‚ã‚Šã¾ã™ï¼ˆå¾Œè¿°ï¼‰ã€‚ ä»Šæœ€é©åŒ–ã§ã®è§£ã®å•é¡Œã¯ãŠã„ã¦ãŠãã¨ã—ã¦ã€ã‚ãã¾ã§è©•ä¾¡ä¸Šã®å•é¡Œãªã‚‰ã€Precision, Recall, ROCæ›²ç·šã€F
ZAORIKU 2019/02/08
æ©Ÿæ¢°å¦ç¿’

è«–æ–‡

@
ãƒªãƒ³ã‚¯
GitHub - zziz/pwc: Papers with code. Sorted by stars. Updated weekly.
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session. Dismiss alert
ZAORIKU 2018/09/11
æ©Ÿæ¢°å¦ç¿’ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ

ã‚ã¨ã§èªã‚€

æ©Ÿæ¢°å¦ç¿’

è«–æ–‡

code

@
ãƒªãƒ³ã‚¯
Jaaxman - arXiv.orgã®äººå·¥çŸ¥èƒ½ã«é–¢ã™ã‚‹è«–æ–‡ã‚’æ—¥æœ¬èªžã§ä¸€è¦§ã§ãã‚‹WEBã‚µã‚¤ãƒˆ
Jaaxmanã¯çµ‚äº†ã„ãŸã—ã¾ã—ãŸ ã”åˆ©ç”¨çŠ¶æ³ã‚’é‘‘ã¿ãŸçµæžœã€èª ã«å‹æ‰‹ãªãŒã‚‰Jaaxmanã¯5æœˆ15æ—¥ã‚’ã‚‚ã£ã¦çµ‚äº†ã„ãŸã—ã¾ã—ãŸã€‚ ã”æ„›é¡§ã„ãŸã ã„ã¦ã„ãŸæ–¹ã€…ã«ã¯å¤§å¤‰å¿ƒè‹¦ã—ã„ã“ã¨ã¨ãªã‚Šæç¸®ã§ã™ãŒã€ã“ã‚Œã¾ã§èª ã«ã‚ã‚ŠãŒã¨ã†ã”ã–ã„ã¾ã—ãŸã€‚
ZAORIKU 2018/06/21
è«–æ–‡

google

ç¿»è¨³

æ©Ÿæ¢°å¦ç¿’

arXiv.org

@
ãƒªãƒ³ã‚¯
ç·šå½¢å±¤ã¨æ¯”è¼ƒã—ãŸç•³ã¿è¾¼ã¿å±¤ - HELLO CYBERNETICS
ã¯ã˜ã‚ã« å…¨çµåˆå±¤ å…¨çµåˆå±¤ã®æ•°å¼ éžå…¨çµåˆå±¤ ç•³ã¿è¾¼ã¿å±¤ å…¥åŠ›æˆåˆ†ã®indexã«ç€ç›®ã—ãŸéžå…¨çµåˆå±¤ éžå…¨çµåˆãƒ»é‡ã¿å…±æœ‰å±¤ 1Dç•³ã¿è¾¼ã¿å±¤ 2Dã®ç•³ã¿è¾¼ã¿å±¤ æœ€å¾Œã« ã¯ã˜ã‚ã« ä»Šå›žã¯ãƒ‹ãƒ¥ãƒ¼ãƒ©ãƒ«ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã®åŸºæœ¬çš„ãªæ§‹æˆè¦ç´ ã§ã‚ã‚‹ç·šå½¢ï¼ˆå…¨çµåˆï¼‰å±¤ã¨ç•³ã¿è¾¼ã¿å±¤ã«ã¤ã„ã¦ç°¡å˜ã«èª¬æ˜Žã—ã¾ã™ã€‚ ã“ã‚Œã‚’èª¬æ˜Žã™ã‚‹ãƒ¢ãƒãƒ™ãƒ¼ã‚·ãƒ§ãƒ³ã¯ã€ã€Œç•³ã¿è¾¼ã¿å±¤ã¯ç·šå½¢å±¤ã‚ˆã‚Šã‚‚å„ªã‚ŒãŸç™ºå±•çš„æ‰‹æ³•ã€è¤‡é›‘ãªæ‰‹æ³•ã§ã‚ã‚‹ã¨ã„ã†å‹˜é•ã„ãŒ åˆå¦è€…ã®ä¸ã§ã‚ã‚‹ã®ã§ã¯ãªã„ã‹ã¨æ„Ÿã˜ã€ãã‚Œã®è§£æ¶ˆã‚’ã™ã‚‹ãŸã‚ã§ã™ã€‚ ä»¥é™ã€ç•³ã¿è¾¼ã¿å±¤ã¨ç·šå½¢å±¤ã®æœ€ãŸã‚‹é•ã„ã‚’ä¸»å¼µã™ã‚‹ãŸã‚ã«ã€ç·šå½¢å±¤ã®ã“ã¨ã‚’å…¨çµåˆå±¤ã¨å‘¼ã¶ã“ã¨ã«ã—ã¾ã™ã€‚ ã“ã®å‘¼ã³åã¯ã€TensorFlowã‚„Kerasãªã©ã®ãƒ•ãƒ¬ãƒ¼ãƒ ãƒ¯ãƒ¼ã‚¯ã§æŽ¡ç”¨ã•ã‚Œã¦ã„ã¾ã™ï¼ˆlayers.denseï¼‰ã€‚ å…¨çµåˆå±¤ å…¨çµåˆå±¤ã®æ•°å¼ ã¾ãšå…¨çµåˆå±¤ã«ã¤ã„ã¦æ•°å¼ã‚’ä»¥ä¸‹ã«ç¤ºã—ã¾ã™ã€‚ å…¥åŠ›ãƒ™ã‚¯ãƒˆãƒ«$x \in \mathbb R^{D}$ã«å¯¾ã—
ZAORIKU 2018/03/11
ãƒ‹ãƒ¥ãƒ¼ãƒ©ãƒ«ãƒãƒƒãƒˆ

nn

@

æ©Ÿæ¢°å¦ç¿’
ãƒªãƒ³ã‚¯
The State of ML and Data Science 2017 | Kaggle
A big picture view of the state of data science and machine learning that shares who is working with data, whatâ€™s happening at the cutting edge of machine learning across industries, and how new data scientists can best break into the field.
ZAORIKU 2017/11/01
science

æ©Ÿæ¢°å¦ç¿’

Machine Learning

@
ãƒªãƒ³ã‚¯
Andrew Ngå…ˆç”Ÿã®æ©Ÿæ¢°å¦ç¿’è¬›ç¾©ã‚’è§£èª¬ã™ã‚‹å‹‰å¼·ä¼šè³‡æ–™ã‚’å…¬é–‹ - ã‚¨ãƒ³ã‚¿ãƒ¼ãƒ—ãƒ©ã‚¤ã‚ºã‚®ãƒ¼ã‚¯ã‚¹ (Enterprise Geeks)
ã‚·ãƒªã‚³ãƒ³ãƒãƒ¬ãƒ¼ã®æœ‰åå¤§å¦ã§ã‚ã‚‹ã‚¹ã‚¿ãƒ³ãƒ•ã‚©ãƒ¼ãƒ‰å¤§å¦ã§ã€Andrew Ngå…ˆç”Ÿã®æ•™ãˆã‚‹æ©Ÿæ¢°å¦ç¿’ã®è¬›ç¾©ãŒäººæ°—ã‚’é›†ã‚ã¦ã„ã‚‹ã€‚ã“ã®è¬›ç¾©ã¯å½¢ã‚’å¤‰ãˆã€courseraã¨ã„ã†ç„¡æ–™ã®Webä¸Šã‚ªãƒ³ãƒ©ã‚¤ãƒ³ã‚³ãƒ¼ã‚¹ã¨ã—ã¦ã‚‚å—è¬›ã§ãã€ã“ã“æ—¥æœ¬ã§ã‚‚æ©Ÿæ¢°å¦ç¿’ã®å‹‰å¼·ãŒã—ãŸã„äººé”ã®é–“ã§ã‚‚äººæ°—ã®è¬›ç¾©ã¨ãªã£ã¦ã„ã‚‹ã€‚ [Machine Learning by Stanford University] https://www.coursera.org/learn/machine-learning/ äººæ°—ã®ç§˜å¯† Andrew Ngå…ˆç”Ÿã®è¬›ç¾©ã¯å†…å®¹ãŒéžå¸¸ã«å®Ÿç”¨çš„ã§ã‚ã‚‹ã“ã¨ãŒäººæ°—ã®ç†ç”±ã®1ã¤ã¨ã—ã¦æŒ™ã’ã‚‰ã‚Œã‚‹ã ã‚ã†ã€‚Andrew Ngå…ˆç”Ÿã¯ã‚·ãƒªã‚³ãƒ³ãƒãƒ¬ãƒ¼ã®ä¼æ¥ã‚’ç›¸æ‰‹ã«æ©Ÿæ¢°å¦ç¿’ã®æŒ‡å°Žã‚’ã—ã¦ã„ã‚‹ã€‚ãŠãã‚‰ããã®çµŒé¨“ã‹ã‚‰ã€ç†è«–ã«èµ°ã‚Šã™ãŽãšã€æ©Ÿæ¢°å¦ç¿’ã‚’é©ç”¨ã™ã‚‹ãƒ—ãƒã‚¸ã‚§ã‚¯ãƒˆã«ãŠã„ã¦æœ¬å½“ã«æœ‰æ„ç¾©ãªå†…å®¹ã«å¼·ããƒ•ã‚©ãƒ¼ã‚«ã‚¹ã—ã¦ã„ã‚‹ã“ã¨ãŒæ„Ÿã˜ã‚‰ã‚Œã‚‹ã®ã ã€‚Andr
ZAORIKU 2017/10/26
MachineLearning

æ©Ÿæ¢°å¦ç¿’

è³‡æ–™

@
ãƒªãƒ³ã‚¯
GitHub - sammy-suyama/BayesBook: ã€Œæ©Ÿæ¢°å¦ç¿’ã‚¹ã‚¿ãƒ¼ãƒˆã‚¢ãƒƒãƒ—ã‚·ãƒªãƒ¼ã‚º ãƒ™ã‚¤ã‚ºæŽ¨è«–ã«ã‚ˆã‚‹æ©Ÿæ¢°å¦ç¿’å…¥é–€ã€ã®ã‚µãƒ³ãƒ—ãƒ«ã‚³ãƒ¼ãƒ‰
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session. Dismiss alert
ZAORIKU 2017/10/20
ãƒ™ã‚¤ã‚º

æ©Ÿæ¢°å¦ç¿’

@

ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°

è³‡æ–™
ãƒªãƒ³ã‚¯
Fantastic GANs and where to find them
This is a series of blog posts. Click here for the second part. Have you ever wanted to know about Generative Adversarial Networks (GANs)? Maybe you just want to catch up on the topic? Or maybe you simply want to see how these networks have been refined over these last years? Well, in these cases, this post might interest you! What this post is not about First things first, this is what you wonâ€™t
ZAORIKU 2017/10/10
GAN

æ©Ÿæ¢°å¦ç¿’

@
ãƒªãƒ³ã‚¯
SIGdial 2017 è«‡è©±å‡¦ç†ãƒ»å¯¾è©±ã‚·ã‚¹ãƒ†ãƒ ã®å‹•å‘ã¾ã¨ã‚ | CyberAgent Developers Blog
ã“ã‚“ã«ã¡ã¯ã€‚ã‚¢ãƒ‰ãƒ†ã‚¯ã‚¹ã‚¿ã‚¸ã‚ª AI Labã®å±±æœ¬ã§ã™ã€‚ è«‡è©±å‡¦ç†ãƒ»å¯¾è©±ã«ã¤ã„ã¦ã®å›½éš›ä¼šè°Â SIGdial 2017Â ã«AI Labã®å±±æœ¬ã¨ç”°ä¸ã§è´è¬›å‚åŠ ã—ã¦ãã¾ã—ãŸã€‚ SIGdialã¯ç”£å¦ã®ç ”ç©¶è€…ãŒæœ€æ–°ã®è«‡è©±ãƒ»å¯¾è©±ç ”ç©¶ã‚’ç™ºè¡¨ã—ã¦ã„ã‚‹å›½éš›ä¼šè°ã§ã™ã€‚ ã¾ãŸã€ä»Šå¹´ã¯æ—¥æœ¬ã‹ã‚‰ã®ç™ºè¡¨è€…ã‚‚æ¯”è¼ƒçš„å¤šãã€æ—¥æœ¬èªžã‚’å¯¾è±¡ã¨ã—ãŸå¯¾è©±ã‚·ã‚¹ãƒ†ãƒ ã®ç ”ç©¶å‹•å‘ã‚‚çŸ¥ã‚‹ã“ã¨ãŒã§ãã¾ã—ãŸã€‚ Openingã®ã‚¹ãƒ©ã‚¤ãƒ‰ã‹ã‚‰ã€‚è‘—è€…ã®å›½åˆ¥çµ±è¨ˆã€‚æ—¥æœ¬ã¯ã‚¢ãƒ¡ãƒªã‚«ã€ã‚¤ã‚®ãƒªã‚¹ã«ç¶šã3ç•ªç›®ã€‚ä»¥ä¸‹ã«ç¶šãã‚«ãƒŠãƒ€ã€ãƒ‰ã‚¤ãƒ„ã¨ã¯åƒ…å·® ã‚¢ãƒ‰ãƒ†ã‚¯ã‚¹ã‚¿ã‚¸ã‚ªã§ã¯ã€ãƒœãƒƒãƒˆå¿œå¯¾ã¨æœ‰äººå¯¾å¿œã‚’èžåˆã•ã›ãŸãƒãƒ£ãƒƒãƒˆãƒœãƒƒãƒˆäº‹æ¥ AI MessengerÂ ã‚’å±•é–‹ã—ã¦ã„ã¾ã™ã€‚ ä»Šå›žã®å¦ä¼šã¯ã€ä»Šå¾Œã©ã®ã‚ˆã†ãªæŠ€è¡“ã‚’å°Žå…¥ã™ã‚Œã°ä»Šã¾ã§ä»¥ä¸Šã«ã‚¯ãƒ©ã‚¤ã‚¢ãƒ³ãƒˆã®æ–¹ã€…ã«åˆ©ç”¨ã—ã¦ã„ãŸã ã‘ã‚‹ã‚µãƒ¼ãƒ“ã‚¹ã«ãªã‚‹ã‹ã‚’è€ƒãˆã‚‹ã‚ˆã„æ©Ÿä¼šã¨ãªã‚Šã¾ã—ãŸã€‚ ã“ã®ãƒ–ãƒã‚°ã§ã¯2äººãŒè´è¬›ã—ãŸç™ºè¡¨ã®ä¸ã‹ã‚‰æ°—ã«ãªã£ãŸå†…å®¹ã‚’ã¾ã¨ã‚ã¦
ZAORIKU 2017/10/08
æ©Ÿæ¢°å¦ç¿’

è«–æ–‡

å¯¾è©±ã‚·ã‚¹ãƒ†ãƒ

AI

@
ãƒªãƒ³ã‚¯
DataScienceCentral.com - Big Data News and Analysis
DSC Weekly 10 January 2023 â€“ Recession Analysis Scott Thompson | January 10, 2023 at 4:42 pm Announcements Recession Analysis Editorâ€™s Note: This weekâ€™s editorial was written by a guest contributor. If... Solution vs. Product-Implications for Agile Development Howard M. Wiener | January 10, 2023 at 2:59 pm A Solution is Not a Product Solution, Deliverable, Product, Work Product and other terms are
ZAORIKU 2017/10/01
æ©Ÿæ¢°å¦ç¿’

ç ”ç©¶

DATA

@
ãƒªãƒ³ã‚¯
lllyasviel.github.io | lllyasviel.github.io
lllyasviel.github.io lllyasviel.github.io ä¸æ–‡ç‰ˆ This is a screenshot of the web APP. If you are a beginner, we recommend you to download the default_refernces.zip. You can get this file by clicking the link marked in red square. Then decompress the zip file and you will get some illustrations like this. There are 15 illustations in the default package with different color. These illustrations are we
ZAORIKU 2017/09/29
ç”»åƒå‡¦ç†

æ©Ÿæ¢°å¦ç¿’

@
ãƒªãƒ³ã‚¯
ãƒ‡ãƒ¼ã‚¿ã«æ¬ æãŒã‚ã‚‹å ´åˆã®æ•™å¸«ã‚ã‚Šå¦ç¿’ - ä½œã£ã¦éŠã¶æ©Ÿæ¢°å¦ç¿’ã€‚
ãŠã¯ã‚ˆã†ã”ã–ã„ã¾ã™ï¼Ž ä»Šå›žã¯æ•™å¸«ã‚ã‚Šå¦ç¿’ãƒ¢ãƒ‡ãƒ«ã‚’é¡Œæã«ï¼Œå…¥åŠ›ãƒ‡ãƒ¼ã‚¿ãŒæ¬ æã—ã¦ã„ã‚‹å ´åˆã®ãƒ™ã‚¤ã‚ºæµã®å¯¾å‡¦æ³•ã‚’è§£èª¬ã—ã¾ã™ï¼Žãƒ™ã‚¤ã‚ºãƒ¢ãƒ‡ãƒ«ã¨ã„ã†ã¨ï¼ŒãŸã„ã¦ã„ã®å ´åˆã¯äº‹å‰åˆ†å¸ƒã®è¨å®šã®ä»•æ–¹äº‘ã€…ã ã¨ã‹ï¼ŒéŽå¦ç¿’ã‚’æŠ‘åˆ¶ã§ãã‚‹ã ã¨ã‹ã«è°è«–ãŒæ³¨ç›®ã•ã‚ŒãŒã¡ã§ã™ãŒï¼Œå€‹äººçš„ã«ã¯ï¼Œãƒ‘ãƒ©ãƒ¡ãƒ¼ã‚¿ã‚„æ½œåœ¨å¤‰æ•°ã‚’æŽ¨è«–ã™ã‚‹ã“ã¨ã¨ã¾ã£ãŸãåŒã˜æž çµ„ã¿ã§æ¬ æå€¤ã‚‚åŒæ™‚ã«æŽ¨è«–ã§ãã‚‹ã“ã¨ãŒï¼Œå®Ÿç”¨ä¸Šéžå¸¸ã«ä¾¿åˆ©ãªãƒ™ã‚¤ã‚ºã®ç‰¹æ€§ã ã¨æ€ã£ã¦ã„ã¾ã™ï¼Ž ãƒ‡ãƒ¼ã‚¿ã®æ¬ æéƒ¨åˆ†ã®å–æ‰±ã„ ãƒ‡ãƒ¼ã‚¿ã«æ¬ æéƒ¨åˆ†ãŒå˜åœ¨ã™ã‚‹ã“ã¨ã¯ã‚ˆãã‚ã‚Šã¾ã™ï¼Žã‚»ãƒ³ã‚µãƒ¼ãƒ‡ãƒ¼ã‚¿ã‚’è§£æžã™ã‚‹éš›ã¯ï¼Œãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã®çŠ¶æ³ã‚„ãƒ‡ãƒã‚¤ã‚¹ã®ä¸å…·åˆã«ã‚ˆã£ã¦ãƒ‡ãƒ¼ã‚¿ã®ä¸€éƒ¨ãŒæ¬ ã‘ãŸçŠ¶æ…‹ã§ä¸ŠãŒã£ã¦ãã‚‹ã“ã¨ãŒã‚ã‚Šã¾ã™ï¼Žã‚¹ãƒžãƒ›ã‹ã‚‰è¤‡æ•°ç¨®é¡žã®ãƒ‡ãƒ¼ã‚¿ã‚’é›†ã‚ã‚‹ã¨ã„ã£ãŸçŠ¶æ³ã‚’è€ƒãˆã¦ã¿ã‚‹ã¨ï¼Œä¾‹ãˆã°åŠ é€Ÿåº¦ã‚»ãƒ³ã‚µãƒ¼ã®å€¤ã¯ç¶™ç¶šçš„ã«å–å¾—ã§ããŸã¨ã—ã¦ã„ã¦ã‚‚ï¼ŒGPSã®ä½ç½®æƒ…å ±ã¯ã»ã¨ã‚“ã©ä¸ŠãŒã£ã¦ã“ãªã„ã¨ã„ã£ãŸå ´åˆã‚‚ã‚ã‚‹ã‹ã¨æ€ã„ã¾ã™ï¼Žã¾ãŸï¼Œä½•ã‹ã—ã‚‰ã®ãƒ¦ãƒ¼
ZAORIKU 2017/09/16
æ©Ÿæ¢°å¦ç¿’

ãƒ™ã‚¤ã‚º

ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°

@
ãƒªãƒ³ã‚¯
My Neural Network isn't working! What should I do?
19/08/2017 So you're developing the next great breakthrough in deep learning but you've hit an unfortunate setback: your neural network isn't working and you have no idea what to do. You go to your boss/supervisor but they don't know either - they are just as new to all of this as you - so what now? Well luckily for you I'm here with a list of all the things you've probably done wrong and compiled
ZAORIKU 2017/09/07
NN

ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°

@

æ©Ÿæ¢°å¦ç¿’

Deep Learning
ãƒªãƒ³ã‚¯
é«˜æ¬¡å…ƒã®ãƒ‡ãƒ¼ã‚¿ã‚’å¯è¦–åŒ–ã™ã‚‹t-SNEã®åŠ¹æžœçš„ãªä½¿ã„æ–¹
t-SNEã¯ã€é«˜æ¬¡å…ƒã®ãƒ‡ãƒ¼ã‚¿ã‚’å¯è¦–åŒ–ã™ã‚‹æ‰‹æ³•ã¨ã—ã¦ã¯ã€éžå¸¸ã«ä¾¿åˆ©ã§ã™ãŒã€æ™‚ã€…ä¸å¯è§£ãªæŒ™å‹•ã‚’ã—ãŸã‚Šã€èª¤è§£ã‚’æ‹›ãã‚ˆã†ãªå¯è¦–åŒ–ã‚’ã™ã‚‹ã“ã¨ãŒã‚ã‚Šã¾ã™ã€‚ ã‚·ãƒ³ãƒ—ãƒ«ãªãƒ‡ãƒ¼ã‚¿ã‚’å¯è¦–åŒ–ã—ã¦å‹•ä½œã®ä»•çµ„ã¿ã‚’ç†è§£ã™ã‚‹ã“ã¨ã§ã€t-SNEã®ã‚ˆã‚ŠåŠ¹æžœçš„ãªä½¿ã„æ–¹ã‚’å¦ã¶ã“ã¨ãŒã§ãã¾ã™ã€‚ t-SNEã¯ã€é«˜æ¬¡å…ƒã®ãƒ‡ãƒ¼ã‚¿ã‚’èª¿æŸ»ã™ã‚‹ãŸã‚ã®æ‰‹æ³•ã¨ã—ã¦ã€2008å¹´ã«van der Maatenã¨Hintonã«ã‚ˆã£ã¦ç™ºè¡¨ [1] ã•ã‚ŒãŸäººæ°—ã®æ‰‹æ³•ã§ã™ã€‚ ã“ã®æŠ€è¡“ã¯ã€æ•°ç™¾ã¾ãŸã¯æ•°åƒæ¬¡å…ƒã®ãƒ‡ãƒ¼ã‚¿ã§ã™ã‚‰ç„¡ç†ã‚„ã‚Š2æ¬¡å…ƒã®ã€Œãƒžãƒƒãƒ—ã€ã«è½ã¨ã—è¾¼ã‚€ã¨ã„ã†ã€ã»ã¨ã‚“ã©é”æ³•ã®ã‚ˆã†ãªèƒ½åŠ›ã‚’å‚™ãˆã¦ã„ã‚‹ãŸã‚ã«ã€æ©Ÿæ¢°å¦ç¿’ã®åˆ†é‡Žã§å¹…åºƒãæ™®åŠã—ã¦ã„ã¾ã™ã€‚ ã“ã®ã‚ˆã†ãªå°è±¡ã‚’æŒã£ã¦ã„ã‚‹æ–¹ãŒå¤šã„ã®ã§ã™ãŒã€ã“ã†ã„ã£ãŸæ‰ãˆæ–¹ã‚’ã—ã¦ã„ã‚‹ã¨èª¤è§£ã‚’æ‹›ãã“ã¨ã‚‚ã‚ã‚Šã¾ã™ã€‚ ã“ã®è¨˜äº‹ã®ç›®çš„ã¯ã€ã‚ˆãã‚ã‚‹å…±é€šã®èª¤è§£ã‚’è§£ããŸã‚ã§ã‚‚ã‚ã‚Šã¾ã™ã€‚ t-SNEã§å¯è¦–åŒ–ã§ãã‚‹ã“ã¨ã¨ã€ã§ããªã„ã“ã¨ã‚’èª¬æ˜Žã™
ZAORIKU 2017/08/28
t-SNE

æ©Ÿæ¢°å¦ç¿’

@
ãƒªãƒ³ã‚¯
Kaggle ç”»åƒåˆ†é¡žã€æ™‚é–“çŸç¸®ã¨ç²¾åº¦å‘ä¸Šã®ãƒ¡ãƒ¢ - ã‚¼ãƒã‹ã‚‰å§‹ã‚ã‚‹æ©Ÿæ¢°å¦ç¿’
Kaggle Planetã«å‚åŠ ã—ã¾ã—ãŸã€‚çµæžœã¯96ä½ã€‚ Discussionã«ãªã„å·¥å¤«ã‚’ã—ãŸç‚¹ã¯å¤©å€™ã«å¿œã˜ã¦ç‰¹å®šã®ãƒ©ãƒ™ãƒ«ã®é–¾å€¤ã‚’å¤‰æ›´ã™ã‚‹ç¨‹åº¦ã§ã€ æå‡ºã—ãŸã®ã¯VGG16, ResNet50, Xceptionã®3CNNã®ã‚¢ãƒ³ã‚µãƒ³ãƒ–ãƒ«ã§ã—ãŸã€‚ æ™‚é–“çŸç¸®ã®ç‚ºã« ç²¾åº¦ã‚’ç«¶ã†ã¨æ©Ÿæ¢°å¦ç¿’ã¯å®Ÿé¨“çš„ãªå´é¢ã‚‚ã‚ã‚Šã€ã„ãã‚‰ã§ã‚‚æ™‚é–“ã‚’æº¶ã‹ã›ã¾ã™ã€‚ Kaggleã®ãƒ©ãƒ³ã‚ãƒ³ã‚°ä¸Šä½ã®ã‚¤ãƒ³ã‚¿ãƒ“ãƒ¥ãƒ¼ã‚’è¦‹ã¦ã‚‚ãƒ‡ãƒ¼ã‚¿ã®ç‰¹å¾´ã‚’ã†ã¾ãæŽ´ã‚€ã“ã¨ãŒå¤§äº‹ã ã¨æ€ã„ã¾ã™ã€‚ ãƒãƒ¼ã‚«ãƒ«ã§ã®æ¤œè¨¼ç’°å¢ƒã‚’æ•´ãˆã‚‹ ãƒ‡ãƒ¼ã‚¿ã®æ•´å½¢ãƒ»æ‹¡å¼µã®æ¯”è¼ƒæ¤œè¨¼ã¯è»½é‡ãªCNNã§ æ™‚é–“ã®ã‹ã‹ã‚‹CNNãƒ¢ãƒ‡ãƒ«ã¯æœ€å¾Œã€ãã®ä»–ã®å‡¦ç†ã«æ³¨åŠ›ã™ã‚‹ æ¡ä»¶ã«ã‚ˆã£ã¦ã‚ã‚‹ç¨‹åº¦æ±ºã‚ã‚Œã‚‹ãƒ‘ãƒ©ãƒ¡ãƒ¼ã‚¿ãƒ¼ã¯å›ºå®šã—ã¦ã—ã¾ã† å¦ç¿’ã¯å¤œé–“ã‚„ä»–ã®ã“ã¨ã‚’ã‚„ã£ã¦ã‚‹æ™‚ã« Discussionã‚„éŽåŽ»ã®å‹è€…ã‚½ãƒªãƒ¥ãƒ¼ã‚·ãƒ§ãƒ³ã‚’è¦‹ã‚‹ ç²¾åº¦å‘ä¸Šã®ç‚ºã« TTAã€€æŽ¨è«–æ™‚ã®ãƒ‡ãƒ¼ã‚¿æ‹¡å¼µ ã‚¢ãƒ³ã‚µãƒ³ãƒ–ãƒ«ã€€å¹³å‡ã€æŠ•ç¥¨ãªã© ã‚³ãƒ¼ãƒ‰ã‚’å†åˆ©ç”¨
ZAORIKU 2017/07/28
æ©Ÿæ¢°å¦ç¿’

kaggle

@

ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°
ãƒªãƒ³ã‚¯
Improving the Realism of Synthetic Images
Most successful examples of neural nets today are trained with supervision. However, to achieve high accuracy, the training sets need to be large, diverse, and accurately annotated, which is costly. An alternative to labelling huge amounts of data is to use synthetic images from a simulator. This is cheap as there is no labeling cost, but the synthetic images may not be realistic enough, resulting
ZAORIKU 2017/07/21
æ©Ÿæ¢°å¦ç¿’

Apple

@

ç ”ç©¶

GAN
ãƒªãƒ³ã‚¯
AI Progress Measurement (Archived)
Measuring the Progress of AI Research (Archived)Â¶ This project was active during 2017 and hasn't been updated since then. It collected probl ems and metrics/datasets from the AI research literature, with the intent to track progress on them. The page is preserved for historical interest but should not be considered up-to-date. You can use this Notebook to see how things are progressing in specific
ZAORIKU 2017/06/22
AI

æ©Ÿæ¢°å¦ç¿’

æŒ‡æ¨™

Python

@

ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°
ãƒªãƒ³ã‚¯
dynet/Autobatching.ipynb at master Â· clab/dynet Â· GitHub
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session.
ZAORIKU 2017/05/23
nn

æ©Ÿæ¢°å¦ç¿’

@

ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°
ãƒªãƒ³ã‚¯
1 2 3 æ¬¡ã®ãƒšãƒ¼ã‚¸