[B! optimization] yu4uã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

yu4u id:yu4u

optimizationã«é–¢ã™ã‚‹yu4uã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (9)

${{author_name}}$

{{{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}}

æœ€é©è¼¸é€ã®è§£ãæ–¹
æœ€é©è¼¸é€å•é¡Œï¼ˆWasserstein è·é›¢ï¼‰ã‚’è§£ãæ–¹æ³•ã«ã¤ã„ã¦ã®ã•ã¾ã–ã¾ãªã‚¢ãƒ—ãƒãƒ¼ãƒãƒ»ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã‚’ç´¹ä»‹ã—ã¾ã™ã€‚ ç·šå½¢è¨ˆç”»ã‚’ä½¿ã£ãŸå®šå¼åŒ–ã®åŸºç¤Žã‹ã‚‰ã¯ã˜ã‚ã¦ã€ä»¥ä¸‹ã®äº”ã¤ã®ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã‚’ç´¹ä»‹ã—ã¾ã™ã€‚ 1. ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚·ãƒ³ãƒ—ãƒ¬ãƒƒã‚¯ã‚¹æ³• 2. ãƒãƒ³ã‚¬ãƒªã‚¢ãƒ³æ³• 3. Sinkhorn ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ 4. ãƒ‹ãƒ¥ãƒ¼ãƒ©ãƒ«ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã«ã‚ˆã‚‹æŽ¨å®š 5. ã‚¹ãƒ©ã‚¤ã‚¹æ³• ã“ã®ã‚¹ãƒ©ã‚¤ãƒ‰ã¯ç¬¬ä¸‰å›ž 0x-seminar https://sites.google.com/view/uda-0x-seminar/home/0x03 ã§ä½¿ç”¨ã—ãŸã‚‚ã®ã§ã™ã€‚è‡ªå·±å®Œçµã™ã‚‹ã‚ˆã†å¿ƒãŒã‘ãŸã®ã§ã‚»ãƒŸãƒŠãƒ¼ã«å‚åŠ ã—ã¦ã„ãªã„äººã«ã‚‚å½¹ç«‹ã¤ã‚¹ãƒ©ã‚¤ãƒ‰ã«ãªã£ã¦ã„ã¾ã™ã€‚ ã€Žæœ€é©è¼¸é€ã®ç†è«–ã¨ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã€å¥½è©•ç™ºå£²ä¸ï¼ https://www.amazon.co.jp/dp/4065305144 Speakerdeck ã«ã‚‚ã‚¢ãƒƒãƒ—ãƒãƒ¼ãƒ‰ã—ã¾ã—ãŸ: https
yu4u 2021/07/01
algorithm

optimization

graph
ãƒªãƒ³ã‚¯
Mathematical Optimization in 60 minutes
æœ¬è¬›æ¼”ã§ã¯ï¼Œæ•°ç†æœ€é©åŒ–ã®åŸºæœ¬çš„ãªæž çµ„ã¿ã‚’æ¦‚è¦³ã™ã‚‹ã“ã¨ã§ï¼Œæ•°ç†æœ€é©åŒ–ã‚’æœ¬æ ¼çš„ã«å¦ç¿’ã™ã‚‹ãã£ã‹ã‘ã‚’ä¸Žãˆã‚‹ã“ã¨ã‚’ç›®çš„ã«ã—ã¦ã„ã¾ã™ï¼Ž ã“ã®ã‚¹ãƒ©ã‚¤ãƒ‰ã§ã¯ï¼ŒåŒå¯¾å•é¡Œã‚’ã¯ã˜ã‚ã¨ã™ã‚‹å¤šãã®é‡è¦ãªæ¦‚å¿µã®èª¬æ˜Žã‚’çœç•¥ã—ã¦ã„ã¾ã™ï¼Žã‚‚ã—ï¼Œã“ã®ã‚¹ãƒ©ã‚¤ãƒ‰ã‚’èªã¿çµ‚ãˆã¦æ•°ç†æœ€é©åŒ–ã‚’æ·±ãç†è§£ã§ããŸã¨æ„Ÿã˜ãŸãªã‚‰ï¼Œãã‚Œã¯ãŸã¶ã‚“æ°—ã®ã›ã„ã§ã™ï¼Žâ€¦
yu4u 2020/06/16
optimization

æœ€é©åŒ–

æ•°ç†æœ€é©åŒ–
ãƒªãƒ³ã‚¯
æœ€é©åŒ–è¶…å…¥é–€
ã™ã†ãŒãå¾’ã®ã¤ã©ã„ã€9/19 ã«ã¦ç™ºè¡¨ãƒãƒ¼ã‚¸ãƒ§ãƒ³ã€‚ è¬›æ¼”å‹•ç”»ã¯ã“ã¡ã‚‰ã€‚ https://www.youtube.com/watch?v=eR3JvijvjW8 ã‚ªãƒªã‚¸ãƒŠãƒ«è«–æ–‡ã¯ã“ã¡ã‚‰ã€‚ https://github.com/kenjihiranabe/The-Art-of-Linear-Algebra/blob/main/The-Art-of-Linear-Algebra.pdf 1ã¤ï¼‘ã¤ã®æŠ€è¡“è§£èª¬ã¯ã“ã¡ã‚‰ã€‚ å„ã€…ã®åˆ†è§£ç‰ã®ãƒˆãƒ”ãƒƒã‚¯ã‚¹è§£èª¬ã¯ã€Qiita ã«è©³ã—ãã—ã¦ã‚ã‚Šã¾ã™ã€‚ https://qiita.com/kenjihiranabe/it ems/bfa9cd68bb355afc56b7
yu4u 2020/05/18
æœ€é©åŒ–

optimization
ãƒªãƒ³ã‚¯
AutoML in NeurIPS 2018
The document discusses two papers presented at NeurIPS 2018 on AutoML: 1) "Massively Parallel Hyperparameter Tuning" which proposes an asynchronous successive halving algorithm to parallelize hyperparameter tuning. This approach improves on synchronous successive halving by avoiding mis-promoting configurations. 2) "Neural Architecture Optimization" which uses a neural network to learn embeddings
yu4u 2019/01/26
automl

optimization
ãƒªãƒ³ã‚¯
POT: Python Optimal Transport â€” Python documentation
POT: Python Optimal TransportÂ¶ The documentation has been moved to : https://PythonOT.github.io
yu4u 2018/10/20
æœ€é©è¼¸é€å•é¡Œã®Pythonã‚½ãƒ«ãƒãƒ©ã‚¤ãƒ–ãƒ©ãƒª

or

optimization

optimal transport
ãƒªãƒ³ã‚¯
Pythonã§Nelder-Meadæ³• - Qiita
Pythonã§Nelder-Meadæ³• åå‰ã®é€šã‚Šã€‚ã ãŒå®Ÿéš›ã¯ Qiitaã‚’ä½¿ã£ã¦ã¿ãŸã‹ã£ãŸ GitHubã‚’ä½¿ã£ã¦ã¿ãŸã‹ã£ãŸ ä»–äººã®ã‚³ãƒ¼ãƒ‰ã‚’è¦‹ã¦å‹‰å¼·ã—ãŸã‹ã£ãŸ ãªã©ã®èƒŒæ™¯ãŒã‚ã‚‹ã®ã§çµæ§‹é›‘ã€‚ Nelder-Meadæ³•ã«ã¤ã„ã¦ æœ€åˆã¯æ”¹è¨‚ã‚·ãƒ³ãƒ—ãƒ¬ãƒƒã‚¯ã‚¹æ³•ã ã¨ã‹æ»‘é™ã‚·ãƒ³ãƒ—ãƒ¬ãƒƒã‚¯ã‚¹æ³•ã ã¨ã‹å‘¼ã°ã‚Œã¦ã„ãŸï¼ˆåŽŸè‘—è«–æ–‡ã«ã‚‚æ›¸ã„ã¦ã‚ã‚‹ã—â€¦ï¼‰ãŒã€wikiã«ã‚‚æ›¸ã‹ã‚Œã¦ã„ã‚‹ã¨ãŠã‚Šã€åŸºæœ¬çš„ã«ã¯ã‚·ãƒ³ãƒ—ãƒ¬ãƒƒã‚¯ã‚¹ã¨ã¯ç„¡é–¢ä¿‚ã€‚ è«–æ–‡ï¼šã€€Nelder, John A.; R. Mead (1965). â€œA simplex method for function minimizationâ€. Computer Journal 7: 308â€“313 è©³ã—ã„ã“ã¨ã¯ï¼ˆè‹±èªžã ãŒï¼‰ã“ã¡ã‚‰ã«åˆ†ã‹ã‚Šã‚„ã™ãã¾ã¨ã‚ã¦ã‚ã‚‹ã€‚ ã–ã£ãã‚Šè¨€ã†ã¨ã€næ¬¡å…ƒã®æœ€é©åŒ–ã—ãŸã„é–¢æ•°ã«n+1æ¬¡å…ƒã®ç‚¹ã®å˜ä½“ï¼ˆsimplexï¼‰ã‚’ã‚³ãƒã‚³ãƒã¨å¤‰å½¢ã•ã›ã¦æœ€å°å€¤ã‚’æŽ¢ã—å‡ºã™ã€‚
yu4u 2018/08/07
optimization
ãƒªãƒ³ã‚¯
Hyperparameter optimization for Neural Networks â€” NeuPy
Hyperparameter optimization for Neural Networks Contents Hyperparameter optimization for Neural Networks Introduction Hyperparameter optimization Grid Search Random Search Hand-tuning Bayesian Optimization Gaussian Process Acquisition Function Find number of hidden units Disadvantages of GP with EI Tree-structured Parzen Estimators (TPE) Overview Hyperparameter optimization for MNIST dataset Disad
yu4u 2017/01/31
optimization

deep learning
ãƒªãƒ³ã‚¯
caffenet-benchmark/batchnorm.md at master Â· ducha-aiki/caffenet-benchmark
This is quick evaluation of BatchNorm layer (BVLC/caffe#3229) performance on ImageNet-2012. Other on-going evaluations: activations [architectures] (https://github.com/ducha-aiki/caffenet-benchmark/blob/master/Architectures.md) [augmentation] (https://github.com/ducha-aiki/caffenet-benchmark/blob/master/Augmentation.md) The architecture is similar to CaffeNet, but has differences: Images are resiz
yu4u 2017/01/25
batch normalizationã¨activationã®é †ç•ªã¨ã‹dropoutã¨ã®çµ„ã¿åˆã‚ã›ã¨ã‹ã®è©•ä¾¡

optimization

tips

deep learning
ãƒªãƒ³ã‚¯
160924 Deep Learning Tuningathon
2. Â© ABEJA, Inc. Deep Learningã®ãƒ‘ãƒ©ãƒ¡ãƒ¼ã‚¿ã‚’ç†è§£ã—ã€ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°ã‚’å®Ÿè·µã™ã‚‹ã“ã¨ã§ Deep Learningã«é–¢ã™ã‚‹ç†è§£ã‚’æ·±ã‚ã‚‹ â– å¦ã¹ã‚‹ã“ã¨ ãƒ»ã©ã®ãƒ‘ãƒ©ãƒ¡ãƒ¼ã‚¿ã‚’å¤‰ãˆã‚‹ã¨å®Ÿè¡Œé€Ÿåº¦ã‚„å¦ç¿’é€Ÿåº¦ãŒã©ã†å¤‰ã‚ã‚‹ã®ã‹ï¼Ÿ ãƒ»ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã®æ§‹ç¯‰æ–¹æ³•ã‚’å¤‰ãˆã‚‹ã¨ã©ã†ç²¾åº¦ã«ã‚¤ãƒ³ãƒ‘ã‚¯ãƒˆã™ã‚‹ã®ã‹ï¼Ÿ â– å¯¾è±¡ Deep Learningã¯ã‚ã‚‹ç¨‹åº¦ä½¿ã£ãŸã“ã¨ã‚ã‚‹ãŒã€ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã®ç´°ã‹ãª ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°ã¯ã©ã†ã™ã‚Œã°ã„ã„ã‹ã‚ã‹ã‚‰ãªã„ ç›®çš„ 3. Â© ABEJA, Inc. Deep Learningã«å¯¾ã™ã‚‹åˆå¦è€…ä»¥ä¸Šã®çŸ¥è˜ Deep Learningã§è‡ªåˆ†ã§ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚’æ›¸ã„ãŸã“ã¨ãŒã‚ã‚‹ Python(numpy, tensorflow)ã«é–¢ã™ã‚‹çŸ¥è˜ Linux, SSH, AWSã«é–¢ã™ã‚‹çŸ¥è˜ ä»Šå›žã®å‹‰å¼·ä¼šã®å¯¾è±¡è€…
yu4u 2016/11/29
deep learning

tips

optimization
ãƒªãƒ³ã‚¯
1