[B! machinelearning] hiroyukimã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

hiroyukim id:hiroyukim

machinelearningã«é–¢ã™ã‚‹hiroyukimã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (14)

${{author_name}}$

{{{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}}

ã€Ruby Kaigi 2017ãƒ¬ãƒãƒ¼ãƒˆã€‘Rubyã§ãƒ‡ãƒ¼ã‚¿ã‚µã‚¤ã‚¨ãƒ³ã‚¹ã¯å‡ºæ¥ã‚‹ã®ã‹ï¼Ÿ
Housmarté«˜æ¾ã§ã™ã€‚ 2017å¹´9æœˆ18-20æ—¥ã®3æ—¥é–“ã€åºƒå³¶ã§é–‹å‚¬ã•ã‚ŒãŸRuby Kaigi 2017ã«å‚åŠ ã—ã¦ãã¾ã—ãŸã€‚Ruby Kaigiè‡ªä½“ã¯åŽ»å¹´ã«å¼•ãç¶šã2å›žç›®ã®å‚åŠ ã¨ãªã‚Šã¾ã™ã€‚ ä»Šå¹´ã‚‚æ§˜ã€…ãªãƒˆãƒ”ãƒƒã‚¯ãŒã‚ã‚Šã¾ã—ãŸãŒã€ä»Šå›žã¯ç§ãŒæ³¨ç›®ã—ã¦ã„ã‚‹ãƒ†ãƒ¼ãƒžã®1ã¤ã§ã‚ã‚‹ â€œRuby x ãƒ‡ãƒ¼ã‚¿ã‚µã‚¤ã‚¨ãƒ³ã‚¹â€ ã®è¦³ç‚¹ã‹ã‚‰Ruby Kaigiã®å‚åŠ ãƒ¬ãƒãƒ¼ãƒˆã‚’ã—ã¾ã™ã€‚ Rubyã‚³ãƒŸãƒ¥ãƒ‹ãƒ†ã‚£ã®ãƒ‡ãƒ¼ã‚¿ã‚µã‚¤ã‚¨ãƒ³ã‚¹ã¸ã®å–ã‚Šçµ„ã¿ ã“ã“2-3å¹´ã§æ©Ÿæ¢°å¦ç¿’ãƒ»ãƒ‡ã‚£ãƒ¼ãƒ—ãƒ©ãƒ¼ãƒ‹ãƒ³ã‚°ã‚„ãƒ‡ãƒ¼ã‚¿ã‚µã‚¤ã‚¨ãƒ³ã‚¹ã‚’æ¥å‹™ã§ä½¿ã†ã“ã¨ã¯ã©ã‚“ã©ã‚“æ™®é€šã®ã“ã¨ã«ãªã£ã¦ãã¦ã„ã¾ã™ã€‚ ã“ã®é ˜åŸŸã¯ã»ã¨ã‚“ã©ãŒPython/R/Scalaãªã©ã§è¡Œã‚ã‚Œã¦ãŠã‚Šã€æœ€è¿‘ã¾ã§Rubyã§ãƒ‡ãƒ¼ã‚¿ã‚µã‚¤ã‚¨ãƒ³ã‚¹ã‚’ã‚„ã‚‹ã®ã¯å®Ÿè³ªä¸å¯èƒ½ãªçŠ¶æ…‹ã§ã—ãŸã€‚ ã§ã™ãŒã€RubyãŒè¨€èªžã¨ã—ã¦ã“ã‚Œã‚‰ã®é ˜åŸŸã‚’ã‚µãƒãƒ¼ãƒˆã™ã‚‹ã‹ã¯ã€Rubyã®ã“ã‚Œã‹ã‚‰ã‚’è€ƒãˆã‚‹æ„å‘³ã§ã‚‚ã¨ã¦ã‚‚é‡è¦ã«ãªã‚Šã¾ã™ã€‚ ã“ã‚Œã‚‰ã®
hiroyukim 2017/09/29
æ©Ÿæ¢°å¦ç¿’

machinelearning

Machine Learning

Python

Ruby

RubyKaigi
ãƒªãƒ³ã‚¯
Sparkï¼MapReduceã® æ©Ÿæ¢°å¦ç¿’ãƒ©ã‚¤ãƒ–ãƒ©ãƒªæ¯”è¼ƒæ¤œè¨¼
2015/5/21 Hadoopã‚½ãƒ¼ã‚¹ã‚³ãƒ¼ãƒ‰ãƒªãƒ¼ãƒ‡ã‚£ãƒ³ã‚° ç¬¬19å›žã«ãŠã‘ã‚‹ãƒªã‚¯ãƒ«ãƒ¼ãƒˆãƒ†ã‚¯ãƒŽãƒã‚¸ãƒ¼ã‚ºå €è¶Šã«ã‚ˆã‚‹ç™ºè¡¨è³‡æ–™ã«ãªã‚Šã¾ã™Read less
hiroyukim 2015/07/06
machinelearning

æ©Ÿæ¢°å¦ç¿’
ãƒªãƒ³ã‚¯
Deep Learning ãƒªãƒ³ã‚¯é›† - äººå·¥çŸ¥èƒ½ã«é–¢ã™ã‚‹æ–å‰µéŒ²
ä¹—ã‚‹ã—ã‹ãªã„ã“ã®ãƒ“ãƒƒã‚°ã‚¦ã‚§ãƒ¼ãƒ–ã« Deep Learningï¼ˆæ·±å±¤å¦ç¿’ï¼‰ã«é–¢é€£ã™ã‚‹ã¾ã¨ã‚ãƒšãƒ¼ã‚¸ã¨ã—ã¦ä½¿ç”¨ã™ã‚‹äºˆå®šã§ã™ã€‚Deep Learningã«é–¢ã™ã‚‹è¨˜äº‹ãƒ»ã‚¹ãƒ©ã‚¤ãƒ‰ãƒ»è«–æ–‡ãƒ»å‹•ç”»ãƒ»æ›¸ç±ã¸ã®ãƒªãƒ³ã‚¯ã‚’ã¾ã¨ã‚ã¦ã„ã¾ã™ã€‚æœ€æ–°ã®ç ”ç©¶å‹•å‘ã¯å…¨ç„¶æŠŠæ¡ã§ãã¦ã„ãªã„ã®ã§ä»Šå¾Œç ”ç©¶ã‚’é€²ã‚ã‚‹ãªã‹ã§è¨˜éŒ²ã—ã¦ã„ããŸã„ã¨æ€ã„ã¾ã™ã€‚èªã‚“ã è«–æ–‡ã®æ¦‚è¦ã‚‚ç°¡å˜ã«ã¾ã¨ã‚ã¦ã„ãäºˆå®šã§ã™ã€‚æœ¬ãƒ–ãƒã‚°ã§ã¯ã€å½“é¢ã®é–“ã€Theanoã‚’ä½¿ã£ã¦å„ç¨®Deep Learningã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã‚’å®Ÿè£…ã—ã¦ã„ããŸã„ã¨æ€ã„ã¾ã™ã€‚ é–¢é€£ãƒ‹ãƒ¥ãƒ¼ã‚¹ãªã©ã¯Twitterã§ã‚‚æµã—ã¦ã„ã‚‹ã®ã§èˆˆå‘³ãŒã‚ã£ãŸã‚‰ãƒ•ã‚©ãƒãƒ¼ã—ã¦ãã ã•ã„ã€‚ ã™ã¹ã¦ã«ç›®ãŒé€šã›ãšæ›´æ–°ãŒè¿½ã„ã¤ã„ã¦ã„ã¾ã›ã‚“ã€‚ç§ã®ã¯ã¦ãªãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ã§[Deep Learning]ã¨ã„ã†ã‚¿ã‚°ã‚’ä»˜ã‘ã¦ç™»éŒ²ã—ã¦ã„ã¾ã™ã€‚ã¾ã£ãŸãæ•´ç†ã§ãã¦ã„ã¾ã›ã‚“ãŒã”å‚è€ƒã¾ã§ã€‚ Theanoç·¨ Theanoã‚’Windowsã«ã‚¤ãƒ³ã‚¹ãƒˆãƒ¼ãƒ«ï¼ˆ2015/1
hiroyukim 2015/05/05
æ©Ÿæ¢°å¦ç¿’

deeplearning

machinelearning

Deep Learning

äººå·¥çŸ¥èƒ½
ãƒªãƒ³ã‚¯
Hivemall talk@Hadoop summit 2014, San Jose
This document discusses Hivemall, a scala ble machine learning library for Apache Hive. It provides concise summaries of machine learning algorithms as user-defined functions that can run on large datasets in Hive. The document outlines the motivation for Hivemall, what algorithms it supports, how to use it to perform tasks like data preparation, training models, and prediction, and how it handles
hiroyukim 2015/04/28
hivemall

machinelearning
ãƒªãƒ³ã‚¯
éžå…¬é–‹ã‚µã‚¤ãƒˆ
ã‚µã‚¤ãƒˆã®æ§‹ç¯‰ã€‚ä½œå“ã®è²©å£²ã€‚ãƒ–ãƒã‚°ã®æŠ•ç¨¿ã€‚ã“ã®ä»–ã«ã‚‚ã•ã¾ã–ã¾ãªæ©Ÿèƒ½ãŒã‚ã‚Šã¾ã™ã€‚ ãƒã‚°ã‚¤ãƒ³ ã‚µã‚¤ãƒˆã‚’ã¯ã˜ã‚ã‚ˆã† éžå…¬é–‹ã‚µã‚¤ãƒˆ ã“ã®ã‚µã‚¤ãƒˆã¯ç¾åœ¨ãƒ—ãƒ©ã‚¤ãƒ™ãƒ¼ãƒˆè¨å®šã«ãªã£ã¦ã„ã¾ã™ã€‚
hiroyukim 2015/03/03
python

machinelearning

Number
ãƒªãƒ³ã‚¯
Learning to Execute
Recurrent Neural Networks (RNNs) with Long Short-Term Memory units (LSTM) are widely used because they are expressive and are easy to train. Our interest lies in empirically evaluating the expressiveness and the learnability of LSTMs in the sequence-to-sequence regime by training them to evaluate short computer programs, a domain that has traditionally been seen as too complex for neural networks.
hiroyukim 2014/10/31
MachineLearning

æ©Ÿæ¢°å¦ç¿’
ãƒªãƒ³ã‚¯
ãƒ‘ã‚¿ãƒ¼ãƒ³èªè˜ãƒ»æ©Ÿæ¢°å¦ç¿’å‹‰å¼·ä¼š
2014å¹´ ãƒ‘ã‚¿ãƒ¼ãƒ³èªè˜ãƒ»æ©Ÿæ¢°å¦ç¿’å‹‰å¼·ä¼š è³‡æ–™ ç¬¬1å›ž:ã‚¤ãƒ³ãƒˆãƒãƒ€ã‚¯ã‚·ãƒ§ãƒ³ ç¬¬2å›ž:ãƒ™ã‚¤ã‚ºç¢ºçŽ‡ãƒ»ãƒ™ã‚¤ã‚ºè˜åˆ¥ãƒ»ãƒ¢ãƒ‡ãƒ«ã®æ¤œè¨¼ ç¬¬3å›ž:ãƒ¢ãƒ‡ãƒ«é¸æŠžåŸºæº–ãƒ»MCMCæ³• ç¬¬4å›ž:ã‚®ãƒ–ã‚¹ã‚µãƒ³ãƒ—ãƒªãƒ³ã‚°ãƒ»ç·šå½¢è˜åˆ¥ãƒ¢ãƒ‡ãƒ« ç¬¬5å›ž:ç·šå½¢è˜åˆ¥ãƒ¢ãƒ‡ãƒ«:ãƒã‚¸ã‚¹ãƒ†ã‚£ãƒƒã‚¯å›žå¸° ç¬¬6å›ž:ç·šå½¢è˜åˆ¥ãƒ¢ãƒ‡ãƒ«:ãƒ™ã‚¤ã‚ºãƒã‚¸ã‚¹ãƒ†ã‚£ãƒƒã‚¯å›žå¸°,ãƒ•ã‚£ãƒƒã‚·ãƒ£ãƒ¼ã®ç·šå½¢åˆ¤åˆ¥ ç¬¬7å›ž:ãƒ‹ãƒ¥ãƒ¼ãƒ©ãƒ«ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ ç¬¬8å›ž:ãƒ‹ãƒ¥ãƒ¼ãƒ©ãƒ«ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯(ç¶šã) ç¬¬9å›ž:ãƒ‹ãƒ¥ãƒ¼ãƒ©ãƒ«ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯(ç¶šã) ç¬¬10å›ž:ã‚«ãƒ¼ãƒãƒ«æ³•, ã‚«ãƒ¼ãƒãƒ«å¯†åº¦æŽ¨å®šæ³•,ã‚«ãƒ¼ãƒãƒ«å›žå¸°åˆ†æž ç¬¬11å›ž:ã‚«ãƒ¼ãƒãƒ«æ³•:ã‚¬ã‚¦ã‚¹éŽç¨‹ ç¬¬12å›ž:ã‚µãƒãƒ¼ãƒˆãƒ™ã‚¯ã‚¿ãƒ¼ãƒžã‚·ãƒ³, ã‚½ãƒ•ãƒˆãƒžãƒ¼ã‚¸ãƒ³ ç¬¬13å›ž:å¤šã‚¯ãƒ©ã‚¹SVM, SVMå›žå¸°, é–¢é€£ãƒ™ã‚¯ã‚¿ãƒ¼ãƒžã‚·ãƒ³ ç¬¬14å›ž:ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ ç¬¬15å›ž:ã‚¸ãƒ§ã‚¤ãƒ³ãƒ„ãƒªãƒ¼ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ç¬¬16å›ž:ã‚¸ãƒ§ã‚¤ãƒ³ãƒ„ãƒªãƒ¼ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ (ç¶šã) ç¬¬17å›ž:ãƒžãƒ«ã‚³ãƒ•ç¢ºçŽ‡å ´ãƒ»ãƒ•ã‚¡ã‚¯ã‚¿ãƒ¼ã‚°ãƒ©
hiroyukim 2014/09/25
æ©Ÿæ¢°å¦ç¿’

MachineLearning

prml
ãƒªãƒ³ã‚¯
å®Ÿè£…ãŒç°¡å˜ã§é«˜æ€§èƒ½ãªç·šå½¢è˜åˆ¥å™¨ã€AdaGrad+RDAã®è§£èª¬ - EchizenBlog-Zwei
æ©Ÿæ¢°å¦ç¿’ã§ã¯ã€ãƒ‡ãƒ¼ã‚¿ãŒã©ã®ã‚¯ãƒ©ã‚¹ã«å±žã™ã‚‹ã‹ã‚’è˜åˆ¥ã™ã‚‹ã¨ã„ã†å•é¡ŒãŒåŸºæœ¬çš„ã§ã™ã€‚ ã“ã®è˜åˆ¥å•é¡Œã¯ç·šå½¢è˜åˆ¥å™¨ã¨ã„ã†ãƒ¢ãƒ‡ãƒ«ã‚’ä½¿ã†ã“ã¨ã§è§£ãã“ã¨ãŒã§ãã¾ã™ã€‚ ã“ã®è¨˜äº‹ã§ã¯ã€å®Ÿè£…ãŒç°¡å˜ã§é«˜æ€§èƒ½ãªç·šå½¢è˜åˆ¥å™¨ã€AdaGrad+RDAã®è§£èª¬ã‚’è¡Œã„ã¾ã™ã€‚ AdaGrad+RDAã®è©³ç´°ãªè§£èª¬ã¯ä»¥ä¸‹ã®è«–æ–‡ã‚’å‚è€ƒã«ã—ã¦ãã ã•ã„ã€‚ http://www.magicbroom.info/Papers/DuchiHaSi10.pdf ã“ã¡ã‚‰ã¯AdaGrad+RDAã®å®Ÿè£…ä¾‹ã§ã™ã€‚ http://d.hatena.ne.jp/echizen_tm/20140726/1406376207 è˜åˆ¥å•é¡Œã¯ã€é€šå¸¸ãƒ‡ãƒ¼ã‚¿ã‚’2ã¤ã®ã‚¯ãƒ©ã‚¹ã«åˆ†é¡žã—ã¾ã™ã€‚ã©ã†ã‚„ã£ã¦åˆ†é¡žã™ã‚‹ã‹ã¨ã„ã†ã¨ã€ç·šå½¢è˜åˆ¥å™¨ãŒæ£ã®å€¤ã‚’è¿”ã—ãŸã‹ã€è² ã®å€¤ã‚’è¿”ã—ãŸã‹ã§ã‚¯ãƒ©ã‚¹ã‚’åˆ†ã‘ã¾ã™ã€‚ å…·ä½“çš„ã«ã¯ã€ç·šå½¢è˜åˆ¥å™¨ã¯ä»¥ä¸‹ã®å½¢å¼ã‚’ã—ã¦ã„ã¾ã™ã€‚ y = Î£(x_i * w_i) ãƒ‡ãƒ¼ã‚¿ã‚’è¡¨
hiroyukim 2014/09/15
MachineLearning
ãƒªãƒ³ã‚¯
AdaGradãŒ12å€é€Ÿããªã‚‹é”æ³•
AdaGradã¯å¦ç¿’çŽ‡ã‚’è‡ªå‹•èª¿æ•´ã—ã¦ãã‚Œã‚‹å‹¾é…æ³•ã®äºœç¨®ã§ã€ã„ã‚ã‚“ãªäººãŒçµ¶è³›ã—ã¦ã„ã¾ã™ã€‚ å‹¾é…ã‚’è¶³ã—è¾¼ã‚€æ™‚ã«ã€å„æ¬¡å…ƒã”ã¨ã«ä»Šã¾ã§ã®å‹¾é…ã®2ä¹—å’Œã‚’ã¨ã£ã¦ãŠã„ã¦ã€ãã®å¹³æ–¹æ ¹ã§å‰²ã£ã¦ã‚ã’ã‚‹ã ã‘ã¨ã€æã‚ã—ãã‚·ãƒ³ãƒ—ãƒ«ã§ã™ã€‚ Adaptive Subgradient Methods for Online Learning and Stochastic Optimization John Duchi, Elad Hazan, Yoram Singer. JMLR 2011. ä¸åº¦ã€ @echizen_tm ã•ã‚“ãŒãƒ–ãƒã‚°ã‚’æ›¸ã„ã¦ã¾ã—ãŸã€‚ AdaGrad+RDAã‚’å®Ÿè£…ã—ã¾ã—ãŸã€‚ é€šå¸¸ã®SGDãªã©ã¯å¦ç¿’çŽ‡ã‚’ã ã‚“ã ã‚“æ¸›è¡°ã•ã›ãªãŒã‚‰å‹¾é…ã‚’è¶³ã—ã¦ã„ãã‚ã‘ã§ã™ãŒã€ã©ã®æ§˜ã«æ¸›è¡°ã•ã›ã‚‹ã‹ã¨ã„ã†å•é¡Œã«ã„ã¤ã‚‚é ã‚’æ‚©ã¾ã›ã¾ã™ã€‚ AdaGradã§ã¯æœ€åˆã®å¦ç¿’çŽ‡ã“ãå¤–ã‹ã‚‰ä¸Žãˆã¾ã™ãŒã€æ¸›è¡°ã®ã•ã›æ–¹ã‚„æ¸›è¡°çŽ‡ã¨ã„ã£ãŸãƒã‚¤ãƒ‘ãƒ¼ãƒ‘ãƒ©ãƒ¡ãƒ¼ã‚¿ã‹ã‚‰
hiroyukim 2014/08/19
Ada

é«˜é€ŸåŒ–

machinelearning
ãƒªãƒ³ã‚¯
åºƒå‘Šã¨æ©Ÿæ¢°å¦ç¿’ - Qiita
Machine Learning Advent Calendarå‘ã‘ã®è¨˜äº‹ã§ã™ã€‚ æ™®æ®µã¯Gunosyã¨ã„ã†ä¼šç¤¾ã§ç¤¾é•·æ¥ã‚’ã—ãªãŒã‚‰ç¤¾é•·ã‚’ã—ã¦ã„ã¾ã™ã€‚ ã–ã£ãã‚Šã„ã†ã¨ çµè«–ã ã‘çŸ¥ã‚ŠãŸã„äººã¯ã“ã“ã ã‘ åºƒå‘Šã«ãŠã‘ã‚‹æ©Ÿæ¢°å¦ç¿’ã®å¿œç”¨ã®å¤šãã¯CTRäºˆæ¸¬ã‚„é‹ç”¨ã®æœ€é©åŒ–ã®ãŸã‚(ã‚¯ã‚¨ãƒªãƒ¼äºˆæ¸¬ã¨ã‹)ã®äºˆæ¸¬å•é¡Œ ä»Šå¾Œã¯ã€ŒCVRã®äºˆæ¸¬ã€ã‚„ã€Œã‚¢ã‚¯ãƒ†ã‚£ãƒ–ãªãƒ¦ãƒ¼ã‚¶ãƒ¼ã®äºˆæ¸¬ã€ãŒãƒ›ãƒƒãƒˆãªè©±é¡Œã«ãªã‚‹(åŠ ãˆã¦ãã®é‹ç”¨ã‚’ã©ã†æœ€é©åŒ–ã™ã‚‹ã‹ã¨ã„ã£ãŸè©±é¡Œã‚‚) ç¾åœ¨ã¯æ¤œç´¢ã‚¨ãƒ³ã‚¸ãƒ³ã®å¿œç”¨ä¾‹ãŒå¤šã„ã€‚ä»Šå¾Œã¯ãƒ‡ã‚£ã‚¹ãƒ—ãƒ¬ã‚¤åºƒå‘Šã‚„ã‚¿ã‚¤ãƒ ãƒ©ã‚¤ãƒ³åºƒå‘Šã¸ã®å¿œç”¨ãŒå¢—ãˆã¦ã„ã å€‹äººã®ãƒ¦ãƒ¼ã‚¶ãƒ¼å±žæ€§ã‚’é›†ã‚ã‚‹ã“ã¨ãŒä»Šã¾ã§ä»¥ä¸Šã«ãƒ¡ãƒ‡ã‚£ã‚¢ã®ãƒ“ã‚¸ãƒã‚¹çš„ã«é‡è¦ã«ãªã‚‹ åºƒå‘Šã‚„æŽ¨è–¦ã‚¨ãƒ³ã‚¸ãƒ³ã«é™ã‚‰ãšãƒ‰ãƒ¡ã‚¤ãƒ³çŸ¥è˜ã¯éžå¸¸ã«é‡è¦ã€‚ãƒ‰ãƒ¡ã‚¤ãƒ³çŸ¥è˜ã¨æ©Ÿæ¢°å¦ç¿’ã®çŸ¥è˜ã‚’æŒã£ãŸã‚¨ãƒ³ã‚¸ãƒ‹ã‚¢ãŒæ„æ€æ±ºå®šã«æºã‚ã‚‹ä¼šç¤¾ã¯ä»Šå¾Œå¤§ããã®ã³ã‚‹(ã¨æ€ã†) åºƒå‘Šã«ã¤ã„ã¦ æœ€è¿‘ã¯ã‚‚ã£ã±ã‚‰åºƒå‘Šã®é–‹ç™ºã‚’ã—ã¦ãŠã‚Šã€åºƒå‘Šåˆ†é‡Žã§
hiroyukim 2014/03/13
æ©Ÿæ¢°å¦ç¿’

machineLearning

åºƒå‘Š
ãƒªãƒ³ã‚¯
è¨ˆç®—è«–çš„ç¥žçµŒç§‘å¦ã¨æ©Ÿæ¢°å¦ç¿’ - Qiita
@torotoki ã§ã™ã€‚Machine Learning Advent Calendar 2012 7æ—¥ç›®ã¯ã€ä¿—ã«è¨€ã†è„³ç§‘å¦ã¨ã‚ˆã°ã‚Œã‚‹ç¥žçµŒç§‘å¦ï¼ˆneuroscienceï¼‰ã®ä¸ã§ã‚‚è¨ˆç®—è«–çš„ç¥žçµŒç§‘å¦ï¼ˆcomputational neuroscienceï¼‰ã¨ã„ã†åˆ†é‡Žã§ã™ã€‚å¦å•çš„ãªå…¥é–€æ›¸ã§ã¯ã¾ãšã€è„³ã®ç¥žçµŒç´°èƒžã®ä»•çµ„ã¿ã‚„ã€å¤šã€…ã®ãƒ‹ãƒ¥ãƒ¼ãƒãƒ³ã®æ•°ç†ãƒ¢ãƒ‡ãƒ«ã‚’å¦ã³ã¾ã™ãŒã€ãã†ã„ã£ãŸã‚‚ã®ã¯å‚è€ƒæ–‡çŒ®ã®è‰¯æ›¸ã«ã¾ã‹ã›ã¦ã€è¨ˆç®—è«–çš„ç¥žçµŒç§‘å¦ã¨ã¯ä½•ã§ã‚ã‚‹ã‹ã€ã¨ã„ã†ã“ã¨ã«ã¤ã„ã¦æ›¸ããŸã„ã¨æ€ã„ã¾ã™ã€‚ è¨ˆç®—è«–çš„ç¥žçµŒç§‘å¦ã¨ã¯ ç¥žçµŒç§‘å¦ã®ç›®æ¨™ã¨ã—ã¦è„³ã®ã—ãã¿ã‚’ç†è§£ã™ã‚‹ã¨ã„ã†ã‚‚ã®ãŒã‚ã‚Šã¾ã™ãŒã€ã©ã†ã—ãŸã‚‰æœ¬å½“ã«è„³ã‚’ç†è§£ã—ãŸã¨è¨€ãˆã‚‹ã®ã‹ã¯é›£ã—ã„å•é¡Œã§ã™ã€‚ã“ã®è¨ˆç®—è«–çš„ç¥žçµŒç§‘å¦ã§ã¯ã€æ•°ç†ãƒ¢ãƒ‡ãƒ«ã‚’ç”¨ã„ã¦ç†è«–çš„ã€ãƒˆãƒƒãƒ—ãƒ€ã‚¦ãƒ³çš„ã«æŽ¨è«–ã—ã€ãã‚Œã‚’å®Ÿéš›ã®è„³ã®æ§‹é€ ã‚„å®Ÿé¨“çµæžœãªã©ãƒœãƒˆãƒ ã‚¢ãƒƒãƒ—çš„çŸ¥è¦‹ã¨å¯¾ç…§ã™ã‚‹ã“ã¨ã«ã‚ˆã‚Šã€è„³ã®ã—ãã¿ã‚’ç†è§£ã—ã‚ˆã†ã¨ã™ã‚‹åˆ†é‡Žã§
hiroyukim 2013/12/16
æ©Ÿæ¢°å¦ç¿’

MachineLearning

machine learning
ãƒªãƒ³ã‚¯
SGD+Î±: ç¢ºçŽ‡çš„å‹¾é…é™ä¸‹æ³•ã®ç¾åœ¨ã¨æœªæ¥
SGDã®æœ€æ–°ã®æ‹¡å¼µæ‰‹æ³•ã‚’ç´¹ä»‹ï¼š Importance-aware Updateã‚„Normalized Online Learningãªã©ï¼ŽSGD+Î±ã¯ã“ã“ã¾ã§å‡ºæ¥ã‚‹ï¼Read less
hiroyukim 2013/10/18
machinelearning

slideshare

æ©Ÿæ¢°å¦ç¿’
ãƒªãƒ³ã‚¯
liblinearã§æ–‡æ›¸åˆ†é¡žã‚’è©¦ã™ - Negative/Positive Thinking
ã¯ã˜ã‚ã« ãƒ‡ãƒ¼ã‚¿æ•´å½¢ã‚„ã‚¹ã‚±ãƒ¼ãƒ«èª¿æ•´ã€ãƒ‘ãƒ©ãƒ¡ãƒ¼ã‚¿ã®æŽ¢ç´¢ã‚’è¡Œã†ã“ã¨ã§ã©ã‚Œã ã‘å¤‰ã‚ã‚‹ã‹æ°—ã«ãªã£ãŸã®ã§ã€liblinearã‚’ä½¿ã£ã¦æ–‡æ›¸åˆ†é¡žã‚’è©¦ã—ã¦ã¿ã‚‹ã€‚ liblinear http://www.csie.ntu.edu.tw/~cjlin/liblinear/ version 1.93ã‚’åˆ©ç”¨ ä½¿ç”¨ã™ã‚‹ãƒ‡ãƒ¼ã‚¿ http://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass.html ã€Œnews20ã€ã‚’ä½¿ç”¨ã™ã‚‹ 20ã‚¯ãƒ©ã‚¹ å¦ç¿’:15935ãƒ‡ãƒ¼ã‚¿ã€ãƒ†ã‚¹ãƒˆ:3993ãƒ‡ãƒ¼ã‚¿ ç´ æ€§æ•°:å¦ç¿’62061ã€ãƒ†ã‚¹ãƒˆ62060 news20.bz2ã¨news20.t.bz2ã¯ã€å˜èªžIDã¨TFå€¤ã®ãƒšã‚¢ã£ã½ã„ #å¦ç¿’ãƒ‡ãƒ¼ã‚¿ã®å„ã‚¯ãƒ©ã‚¹ã®ãƒ‰ã‚ãƒ¥ãƒ¡ãƒ³ãƒˆæ•° $ cut -f1 -d" " news20 | sort |uniq -c | sort -k2 -n
hiroyukim 2013/09/01
MachineLearning
ãƒªãƒ³ã‚¯
Publications
Here is a list of my publications, together with abstracts, bibtex entries and pdfs. The copyrights belong to the publishers, so the papers are offered here only for personal use on a non-commercial basis. More formally, this work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License No pdfs are given for books and some book chapters once they appea
hiroyukim 2013/07/17
machinelearning

data structure
ãƒªãƒ³ã‚¯
1