[B! machine learning] johshishaã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

johshisha id:johshisha

machine learningã«é–¢ã™ã‚‹johshishaã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (8)

${{author_name}}$

{{{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}}

conditional DRAGANã§ã®ãƒ©ãƒ™ãƒ«ã®ä¸Žãˆæ–¹ - ç·‘èŒ¶æ€è€ƒãƒ–ãƒã‚°
conditional GANã®ãƒ©ãƒ™ãƒ«ã®ä¸Žãˆæ–¹ã¯è‰²ã€…ã‚ã‚Šã€ æ¯Žå›žã©ã†ã™ã‚Œã°è‰¯ã„ã‹ã‚ˆãã‚ã‹ã‚‰ãšè¿·ã£ã¦ã—ã¾ã†ã€‚ githubã®å®Ÿè£…ã‚’ã¿ã‚‹ã¨æ§˜ã€…ã«æ›¸ã‹ã‚Œã¦ã„ã‚‹ã€‚ æ–‡çŒ®ã‚’èªã‚€ã‚ˆã‚Šã‚‚è‰²ã‚“ãªäººã®å®Ÿè£…ã‚’æ¼ã‚‹ã»ã†ãŒçŸ¥è¦‹ãŒè²¯ã¾ã‚‹ã“ã®é ƒã€‚ ä»Šå›žã¯MNISTã«å¯¾ã—ã¦DRAGANã‚’ç”¨ã„ã¦ã€ ãã®ä¸ã§ã‚‚ã‚ˆãè¦‹ã‹ã‘ã‚‹è¨å®šã‚’è©•ä¾¡ã—ã¦ã¿ã‚‹ã€‚ ï¼ˆDRAGANã‚’ç”¨ã„ãŸã®ã¯å˜ã«å®‰å®šã—ã¦ã„ã‚‹ã¨ã„ã†ç†ç”±ã®ã¿ï¼‰ ä»¥ä¸‹ã®è¨å®šã®ã‚‚ã¨ã§ã„ãšã‚Œã®è©•ä¾¡ã‚‚è¡Œã£ãŸï¼Ž batchsizeã¯64 optimizerã¯Adamã‚’ä½¿ç”¨ï¼ˆlearning rate:1e-4ï¼‰ Generatorã®æ§‹é€ ã¯3å±¤(1:FCå±¤, 2:Deconvå±¤, 3:Deconvå±¤)ï¼Œ å‡ºåŠ›å±¤ã®æ´»æ€§åŒ–é–¢æ•°ã¯tanhã§ã¯ãªãsigmoidã‚’ä½¿ç”¨ Discriminatorã®æ§‹é€ ã¯3å±¤(1:Convå±¤, 2:Convå±¤, 3: FCå±¤) BatchNormalizationä¸ä½¿ç”¨
johshisha 2017/10/25
ã‚ã¨ã§èªã‚€

GAN

machine learning

deep learning
ãƒªãƒ³ã‚¯
https://rll.berkeley.edu/deeprlcourse/
johshisha 2017/07/07
å¼·åŒ–å¦ç¿’

æ©Ÿæ¢°å¦ç¿’

reinforcement learning

machine learning

deep learning
ãƒªãƒ³ã‚¯
æ©Ÿæ¢°å¦ç¿’ãƒ¢ãƒ‡ãƒ«ã®å®Ÿè£…ã«ãŠã‘ã‚‹ã€ãƒ†ã‚¹ãƒˆã«ã¤ã„ã¦ - Qiita
æ•°ã‚ã‚‹ãƒ•ãƒ¬ãƒ¼ãƒ ãƒ¯ãƒ¼ã‚¯ã«ä»˜å±žã™ã‚‹Exampleã€æ©Ÿæ¢°å¦ç¿’ãƒ¢ãƒ‡ãƒ«ã‚’å®Ÿè£…ã—ã¦ã¿ãŸã€ã¨ã„ã†è©±ã€‚ã“ã‚Œã‚‰ã«å…±é€šã—ã¦è¨€ãˆã‚‹ã®ã¯ã€Œãƒ†ã‚¹ãƒˆãŒãªã„ã€ã¨ã„ã†ã“ã¨ã§ã™ã€‚ æ©Ÿæ¢°å¦ç¿’ã®ãƒ¢ãƒ‡ãƒ«ã ã£ã¦ã€ã‚¢ãƒ—ãƒªã‚±ãƒ¼ã‚·ãƒ§ãƒ³ã«çµ„ã¿è¾¼ã¾ã‚Œã‚Œã°ãƒ—ãƒãƒ€ã‚¯ã‚·ãƒ§ãƒ³ã‚³ãƒ¼ãƒ‰ã®ä¸€éƒ¨ã§ã™ã€‚ãƒ†ã‚¹ãƒˆãŒãªã„å®Ÿè£…ã‚’æœ¬ç•ªç’°å¢ƒã«çµ„ã¿è¾¼ã‚€ã‹ï¼Ÿã¨ã„ã†ã¨é€šå¸¸ãã‚“ãªã“ã¨ã‚ã‚Šãˆãªã„ã¨æ€ã„ã¾ã™ã€‚ ([ã‚¹ã‚¿ã‚¸ã‚ªã‚¸ãƒ–ãƒª ç´…ã®è±š](https://www.amazon.co.jp/dp/B00005R5J6) ã‚ˆã‚Šæ‹å€Ÿ) å¿˜ã‚Œã‚‰ã‚ŒãŒã¡ãªç‚¹ã§ã™ãŒã€æ©Ÿæ¢°å¦ç¿’ãƒ¢ãƒ‡ãƒ«ã¯ã€Œãƒªãƒªãƒ¼ã‚¹ã—ãŸçž¬é–“ã€ãŒæœ€é«˜ç²¾åº¦ã«ãªã‚Šã¾ã™ã€‚ãªãœãªã‚‰ã€ãƒªãƒªãƒ¼ã‚¹ã—ãŸçž¬é–“ã“ããŒãã®æ™‚ç‚¹ã§æ‰‹ã«å…¥ã‚‹ãƒ•ãƒ«ãƒ•ãƒ«ã®ãƒ‡ãƒ¼ã‚¿ã‚’ä½¿ã£ã¦é›ãˆä¸Šã’ãŸãƒ¢ãƒ‡ãƒ«ã§ã‚ã‚Šã€ãã‚Œä»¥é™ã¯ã©ã‚“ã©ã‚“æœªçŸ¥ã®ãƒ‡ãƒ¼ã‚¿ãŒå…¥ã£ã¦ãã‚‹ãŸã‚ã§ã™ã€‚ ãã®ãŸã‚ã€ãƒ¢ãƒ‡ãƒ«ã®ç²¾åº¦ã€ã¾ãŸå¦¥å½“æ€§ã‚’ã„ã¤ã§ã‚‚æ¤œè¨¼ã§ãã‚‹ã‚ˆã†ã«ã—ã¦ãŠãã¨ã„ã†ã®ã¯ã¨ã¦ã‚‚é‡è¦ã§ã™ã€‚ã“ã‚Œã¯é€šå¸¸ã®ã‚³ãƒ¼ãƒ‰ã«ãƒ†ã‚¹ãƒˆã‚’ã¤
johshisha 2017/06/25
æ©Ÿæ¢°å¦ç¿’

ãƒ†ã‚¹ãƒˆ

è¨è¨ˆ

deep learning

machine learning
ãƒªãƒ³ã‚¯
7 Techniques to Handle Imbalanced Data - KDNuggets
7 Techniques to Handle Imbalanced Data This blog post introduces seven techniques that are commonly applied in domains like intrusion detection or real-time bidding, because the datasets are often extremely imbalanced. By Ye Wu & Rick Radewagen Introduction What have datasets in domains like, fraud detection in banking, real-time bidding in marketingÂ or intrusion detection in networks, in common?
johshisha 2017/06/10
ãƒ‡ãƒ¼ã‚¿ä¸å‡è¡¡

ã‚ã¨ã§èªã‚€

æ©Ÿæ¢°å¦ç¿’

machine learning
ãƒªãƒ³ã‚¯
The Marginal Value of Adaptive Gradient Methods in Machine Learning
Adaptive optimization methods, which perform local optimization with a metric constructed from the history of iterates, are becoming increasingly popular for training deep neural networks. Examples include AdaGrad, RMSProp, and Adam. We show that for simple overparameterized probl ems, adaptive methods often find drastically different solutions than gradient descent (GD) or stochastic gradient desc
johshisha 2017/06/02
adamã‚ˆã‚ŠSGDã®ã»ã†ãŒã‚ˆã„ã‚‰ã—ã„

machine learning
ãƒªãƒ³ã‚¯
An Overview of Multi-Task Learning for Deep Learning
An Overview of Multi-Task Learning in Deep Neural Networks Multi-task learning is becoming more and more popular. This post gives a general overview of the current state of multi-task learning. In particular, it provides context for current neural network-based methods by discussing the extensive multi-task learning literature. Note: If you are looking for a review paper, this blog post is also av
johshisha 2017/06/02
ãƒžãƒ«ãƒã‚¿ã‚¹ã‚¯ ã‚µãƒ¼ãƒ™ã‚¤

deep learning

machine learning
ãƒªãƒ³ã‚¯
æ©Ÿæ¢°å¦ç¿’ ã¯ã˜ã‚ã‚ˆã† è¨˜äº‹ä¸€è¦§ | gihyo.jp
é‹å–¶å…ƒã®ãƒã‚´ Copyright Â© 2007-2024 All Rights Reserved by Gijutsu-Hyoron Co., Ltd. ãƒšãƒ¼ã‚¸å†…å®¹ã®å…¨éƒ¨ã‚ã‚‹ã„ã¯ä¸€éƒ¨ã‚’ç„¡æ–ã§åˆ©ç”¨ã™ã‚‹ã“ã¨ã‚’ç¦æ¢ã—ã¾ã™â ã€‚å€‹åˆ¥ã«ãƒ©ã‚¤ã‚»ãƒ³ã‚¹ãŒè¨å®šã•ã‚Œã¦ã„ã‚‹è¨˜äº‹ç‰ã¯ãã®ãƒ©ã‚¤ã‚»ãƒ³ã‚¹ã«å¾“ã„ã¾ã™ã€‚
johshisha 2017/06/01
ã‚ã¨ã§èªã‚€

æ©Ÿæ¢°å¦ç¿’

machine learning
ãƒªãƒ³ã‚¯
æ©Ÿæ¢°å¦ç¿’ã‚¨ãƒ³ã‚¸ãƒ‹ã‚¢ãŒçŸ¥ã‚‹ã¹ã10ã®ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ
å‰å›žã®æŠ•ç¨¿ã‹ã‚‰å¤§åˆ†æ™‚é–“ãŒç©ºã„ã¦ã—ã¾ã„ã¾ã—ãŸãŒã€ç¾åœ¨ãƒžãƒ¼ã‚±ãƒ†ã‚£ãƒ³ã‚°éƒ¨ã§ã¯ã€ãƒ‡ãƒ¼ã‚¿ã‚µã‚¤ã‚¨ãƒ³ã‚¹ã«é–¢ã™ã‚‹çŸ¥è˜ã‚’æ·±ã‚ã‚‹ãŸã‚ã«æµ·å¤–ã®ãƒ‡ãƒ¼ã‚¿ã‚µã‚¤ã‚¨ãƒ³ã‚¹è¨˜äº‹ã‚’ç¿»è¨³ã™ã‚‹ã¨ã„ã†å–ã‚Šçµ„ã¿ã‚’è¡Œã£ã¦ã„ã¾ã™ã€‚ä¸»ã«ã€KDnuggetsã¨ã„ã†ã‚µã‚¤ãƒˆã§ç´¹ä»‹ã•ã‚Œã¦ã„ã‚‹è¨˜äº‹ã§äººæ°—ã®ã‚ã‚‹ã‚‚ã®ã‚’ä¸å¿ƒã«é¸ã³ã€åŽŸè‘—è€…ã‚ˆã‚Šç¿»è¨³ã®è¨±è«¾ã‚’ã„ãŸã ã‘ãŸè¨˜äº‹ã‚’å…¬é–‹ã—ã¦ãŠã‚Šã¾ã™ã€‚ä¸å®šæœŸã§ã™ãŒã€ãƒ‡ãƒ¼ã‚¿ã‚µã‚¤ã‚¨ãƒ³ã‚¹ã«é–¢å¿ƒã®ã‚ã‚‹çš†æ§˜ã«ã‚ˆã‚Šè‰¯ã„æƒ…å ±ã‚’æ—¥æœ¬èªžã§ãŠå±Šã‘ã§ãã‚‹ã‚ˆã†ã«å–ã‚Šçµ„ã‚“ã§å‚ã‚Šã¾ã™ã€‚ åˆå›žã«å–ã‚Šä¸Šã’ãŸã„è¨˜äº‹ã¯James Leæ°ã®ã€ŒThe 10 Algorithms Machine Learning Engineers Need to Knowã€ã§ã™ã€‚æ©Ÿæ¢°å¦ç¿’ã®æ‰‹æ³•ãŒç¶²ç¾…çš„ã«ç´¹ä»‹ã•ã‚Œã¦ãŠã‚Šã€å®Ÿç”¨ä¾‹ã‚‚ç¤ºã•ã‚Œã¦ã„ã¾ã™ã€‚åˆã‚ã¦æ©Ÿæ¢°å¦ç¿’ã«å–ã‚Šçµ„ã‚€æ–¹ã®ã”å‚è€ƒã«ãªã‚Œã°å¹¸ã„ã§ã™ã€‚ SOURCE https://gab41.lab41.org/the-10-alg
johshisha 2017/03/10
ã‚ã¨ã§èªã‚€

æ©Ÿæ¢°å¦ç¿’

machine learning
ãƒªãƒ³ã‚¯
1