Nonparametric Bayes Tutorial
Machine Learning Summer School 2012 Bayesian nonparametrics. P Orbanz. Machine Learning Summer School, 2012. [Videolectures] [Slides] Lecture 1: Clustering, Dirichlet processes, IBPs [Slides] Lecture 2: Gaussian processes, model construction, exchangeability, asymptotics Machine Learning Summer School 2009 At MLSS 2009, I gave two talks on the basics of measure theory and stochastic process concep
Spectral Approaches to Learning Latent Variable Models: ICML 2012
Slides, sample code, and videos are now available (see below on this page). Many problems in machine learning involve collecting high-dimensional multivariate observations or sequences of observations, and then hypothesizing a compact model which explains these observations. An appealing representation for such a model is a latent variable model that relates a set of observed variables to an addit
Deep Learning Tutorials — DeepLearning 0.1 documentation
Deep Learning Tutorials¶ Deep Learning is a new area of Machine Learning research, which has been introduced with the objective of moving Machine Learning closer to one of its original goals: Artificial Intelligence. See these course notes for a brief introduction to Machine Learning for AI and an introduction to Deep Learning algorithms. Deep Learning is about learning multiple levels of represen
representation-learning-tutorial-2012
ICML 2012 Representation Learning Tutorial Yoshua Bengio U. Montreal June 26th, 2012, Edinburgh, Scotland The success of machine learning algorithms generally depends on data representation, and we hypothesize that this is because different representations can entangle and hide more or less the different explanatory factors of variation behind the data. Although domain knowledge can be used to he
K-Fold Cross-Validation
SLDM III c Hastie & Tibshirani - February 25, 2009 Cross-validation and bootstrap 3 K-Fold Cross-Validation Primary method for estimating a tuning parameter λ (such as subset size) • Divide the data into K roughly equal parts 1 TrainTrainValidation Train Train 5432 • for each k = 1, 2, . . . K, ï¬t the model with parameter λ to the other K − 1 parts, giving ˆβ−k (λ) and compute its error in predic
Software by Radford Neal Available On-Line
The following software packages are available for free use. All use of this software is at the user's own risk. pqR - my version of the R interpreter, with improved performance and extensions, including a preliminary version of automatic differentiation. Software for flexible Bayesian modeling and Markov chain sampling. This software implements Bayesian models for regression and classification bas
å”調フィルタリングã®ã‚°ãƒ©ãƒ•ã‚£ã‚«ãƒ«ãƒ¢ãƒ‡ãƒ« - nokunoã®æ—¥è¨˜
å”調フィルタリングã¨ã¯Amazonã®ãŠå‹§ã‚ã®ã‚ˆã†ã«ã€Œã“ã®å•†å“を購入ã—ãŸäººã¯ã“ã‚“ãªå•†å“も購入ã—ã¦ã„ã¾ã™ã€ã¨ã„ã†æƒ…å ±ã‚’ç”¨ã„ã¦æŽ¨è–¦ã‚’ã™ã‚‹æ‰‹æ³•ã§ã™ã€‚グラフィカルモデルã¯ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã¨ã‚‚呼ã°ã‚Œã€æœ€è¿‘一部ã§æµè¡Œã—ã¦ã„る機械å¦ç¿’ã®æ‰‹æ³•ã§ã™ã€‚今回ã¯ã€å”調フィルタリングをグラフィカルモデルã§è¡¨ç¾ã—ãŸã‚‰ã©ã®ã‚ˆã†ã«ãªã‚‹ã ã‚ã†ã€ã¨è€ƒãˆã¦æ€ã„ã¤ã„ãŸã‚¢ã‚¤ãƒ‡ã‚¢ã‚’紹介ã—ã¾ã™ã€‚ 今ã€ãƒ¦ãƒ¼ã‚¶uã¨ã‚¢ã‚¤ãƒ†ãƒ iã®çµ„{u,i}ã®ãƒ‡ãƒ¼ã‚¿ãŒå¤§é‡ã«ä¸Žãˆã‚‰ã‚Œã¦ã„ã‚‹ã¨ã—ã¾ã™ã€‚例ãˆã°ã‚½ãƒ¼ã‚·ãƒ£ãƒ«ãƒ–ックマークãªã‚‰ãƒ¦ãƒ¼ã‚¶ã¨ãƒ–ックマークã—ã¦ã„るページã®çµ„ã¿åˆã‚ã›ã€E-commerseãªã‚‰ãƒ¦ãƒ¼ã‚¶ã¨è³¼å…¥ã—ãŸå•†å“ã®çµ„ã¿åˆã‚ã›ã€ãªã©ã§ã™ã€‚ã“ã“ã§ã¯SBMを例ã«è€ƒãˆã‚‹ã®ã§ã€ã¯ã¦ãƒ–ã¨åŒæ§˜ã«ãƒ¦ãƒ¼ã‚¶ã¯ãƒžã‚¤ãƒŠã‚¹ã®è©•ä¾¡ã‚’付ã‘ã‚‹ã“ã¨ã¯ã§ããªã„ã‚‚ã®ã¨ã—ã¾ã™ã€‚ ã“ã®ã¨ãユーザuã«å¯¾ã—ã¦ãŠå‹§ã‚ã®ãƒšãƒ¼ã‚¸ã‚’推薦ã™ã‚‹ã“ã¨ã‚’考ãˆã‚‹ã¨ã€ãƒ¦ãƒ¼ã‚¶uãŒã¾ã ブックマークã—ã¦ã„ãªã„ページiã«
Hashing with Graphs - Anchor Graph Hashing
1. The document describes Anchor Graph Hashing (AGH), a method for learning binary codes for approximate nearest neighbor search using graphs. 2. AGH constructs an anchor graph from a set of anchor points and learns binary codes by solving a graph partitioning problem on the anchor graph. 3. AGH has time and space complexities that are sublinear in the number of data points for training and effici
次世代シーケンサãŒæ±‚ã‚る機械å¦ç¿’
2014.06.08 「ベイズ推定ã«ã‚ˆã‚‹å¤šå¤‰é‡è§£æžå…¥é–€ã€ï¼ åºƒå³¶å¤§å¦ ãƒ™ã‚¤ã‚ºæŽ¨å®šã¨ãƒžãƒ«ã‚³ãƒ•é€£éŽ–モンテカルãƒæ³•ã«ã¤ã„ã¦è§£èª¬ã—ã¦ã„ã¾ã™ã€‚ 当日動ã‹ã—ãŸã‚®ãƒ–スサンプリングã®ã‚¢ãƒ‹ãƒ¡ãƒ¼ã‚·ãƒ§ãƒ³ã¯ï¼Œ 以下ã®ã‚µã‚¤ãƒˆã§ç´¹ä»‹ã•ã‚ŒãŸã‚³ãƒ¼ãƒ‰ã‚’å‚考ã«ã—ã¦ã„ã¾ã™ã€‚ http://qiita.com/yagays/items/5bde6addf228b1fe24e6
機械å¦ç¿’ã«ãŠã‘る連続最é©åŒ–ã®æ–°ã—ã„トレンド | PDF
機械å¦ç¿’ã§ã¯æ—¢å˜ã®æœ€é©åŒ–手法をãŸã æŒã£ã¦ãã¦é©ç”¨ã™ã‚Œã°ã‚ˆã„ã‹ã¨ã„ã†ã¨ï¼Œæœ€è¿‘ã®ãƒ‡ãƒ¼ã‚¿ã‚µã‚¤ã‚ºã®å¢—åŠ ã¨ã¨ã‚‚ã«ãã®ã‚ˆã†ãªç°¡å˜ãªçŠ¶æ³ã§ã¯ãªããªã£ã¦ã„ã¾ã™ï¼Žå¤§ããªãƒ‡ãƒ¼ã‚¿ã‚’解æžã™ã‚‹ãŸã‚ã«ã¯ã‚ã¾ã‚Šè¤‡é›‘ãªæ‰‹æ³•ã¯ä½¿ãˆãªã„代ã‚ã‚Šã«å•é¡Œã®æ€§è³ªã‚’æ·±ã考慮ã™ã‚‹ã“ã¨ãŒéžå¸¸ã«é‡è¦ã§ã™ï¼Žãã®ãŸã‚，内点法ãªã©ã®æœ€æ–°ã®æœ€é©åŒ–手法ã§ã¯ãªã，60年代-70年代ã«ç ”究ã•ã‚ŒãŸå¤ã„手法ã«æ³¨ç›®ãŒé›†ã¾ã£ã¦ã„ã¾ã™ï¼Žã“ã®ç™ºè¡¨ã§ã¯ã‚¹ãƒ‘… Document Informationclick to expand document information機械å¦ç¿’ã§ã¯æ—¢å˜ã®æœ€é©åŒ–手法をãŸã æŒã£ã¦ãã¦é©ç”¨ã™ã‚Œã°ã‚ˆã„ã‹ã¨ã„ã†ã¨ï¼Œæœ€è¿‘ã®ãƒ‡ãƒ¼ã‚¿ã‚µã‚¤ã‚ºã®å¢—åŠ ã¨ã¨ã‚‚ã«ãã®ã‚ˆã†ãªç°¡å˜ãªçŠ¶æ³ã§ã¯ãªããªã£ã¦ã„ã¾ã™ï¼Žå¤§ããªãƒ‡ãƒ¼ã‚¿ã‚’解æžã™ã‚‹ãŸã‚ã«ã¯ã‚ã¾ã‚Šè¤‡é›‘ãªæ‰‹æ³•ã¯ä½¿ãˆãªã„代ã‚ã‚Šã«å•é¡Œã®æ€§è³ªã‚’æ·±ã考慮ã™ã‚‹ã“ã¨ãŒéžå¸¸ã«é‡è¦ã§ã™ï¼Žãã®ãŸã‚，内点法ãªã©ã®æœ€æ–°ã®æœ€é©åŒ–手法ã§ã¯ãªã，6
é¢ç™½ãã†ã¨æ€ã„ã¤ã¤èªã‚ã¦ãªã„è«–æ–‡é” - yasuhisa's blog
自分用メモ。今ã¯èªã‚“ã§ã‚‹å ´åˆã˜ã‚ƒãªã„ã®ã ã‘ã‚Œã©(誰ã‹ç´¹ä»‹ã—ã¦)ã€ã©ã£ã‹ã«ãƒ¡ãƒ¢ã£ã¨ã‹ãªã„ã¨å¿˜ã‚Œã‚‹è«–文。"ã‹ãªã‚Š"é‡è¦ãã†ãªã®ã‚’ピックアップã—ã¦ã‚る。 Painless Unsupervised Learning with Features (pdf, presentation) Parsing Natural Scenes and Natural Language with Recursive Neural Networks (pdf) Joint Parsing and Alignment with Weakly Synchronized Grammars (pdf, presentation) Type-Based MCMC (pdf, presentation) Distance Dependent Chinese Restaurant Processes (pdf) ã¨ã„ã†ã‹ã¾ã èªã‚“
ICML2011èªã¿ä¼šã‚’開催ã—ã¾ã—㟠#icmlreading - nokunoã®æ—¥è¨˜
ã¨ã„ã†ã‚ã‘ã§PFIã§é–‹å‚¬ã—ã¾ã—ãŸ!ä¼èª¬ã®ãƒ¬ãƒƒãƒ‰ãƒ–ル倉庫ã¯ã“ã¡ã‚‰ï¼ŽICMLã¯æ©Ÿæ¢°å¦ç¿’分野ã®ãƒˆãƒƒãƒ—カンファレンスã§ã™ï¼Žä»Šå›žã¯ï¼ŒICML2011ã®äºˆç¨¿é›†ã‹ã‚‰1人30分程度ã§è«–æ–‡ã®ç´¹ä»‹ã‚’è¡Œã„ã¾ã—ãŸï¼ŽICML 2011, The 28th International Conference on Machine Learning - Bellevue, Washington Sparse Additive Generative Model for Text by @tsubosakaã•ã‚“Icml2011 reading-sage View more presentations from tsubosakaè«–æ–‡ 背景 LDAãªã©ã®Dirichlet-Multinomialを使ã£ãŸæ–‡æ›¸ç”Ÿæˆãƒ¢ãƒ‡ãƒ«ãŒæµè¡Œã£ã¦ã„ã‚‹ å•é¡Œ1：Inference Cost å•é¡Œ2：Over parametrization å•é¡Œ3：
クラスタリング (クラスター分æž) - Toshihiro Kamishima
クラスタリング (clustering) ã¨ã¯ï¼Œåˆ†é¡žå¯¾è±¡ã®é›†åˆã‚’，内的çµåˆ (internal cohesion) ã¨å¤–的分離 (external isolation) ãŒé”æˆã•ã‚Œã‚‹ã‚ˆã†ãªéƒ¨åˆ†é›†åˆã«åˆ†å‰²ã™ã‚‹ã“㨠[Everitt 93, 大橋 85] ã§ã™ï¼Žçµ±è¨ˆè§£æžã‚„多変é‡è§£æžã®åˆ†é‡Žã§ã¯ã‚¯ãƒ©ã‚¹ã‚¿ãƒ¼åˆ†æž (cluster analysis) ã¨ã‚‚呼ã°ã‚Œï¼ŒåŸºæœ¬çš„ãªãƒ‡ãƒ¼ã‚¿è§£æžæ‰‹æ³•ã¨ã—ã¦ãƒ‡ãƒ¼ã‚¿ãƒžã‚¤ãƒ‹ãƒ³ã‚°ã§ã‚‚é »ç¹ã«åˆ©ç”¨ã•ã‚Œã¦ã„ã¾ã™ï¼Ž 分割後ã®å„部分集åˆã¯ã‚¯ãƒ©ã‚¹ã‚¿ã¨å‘¼ã°ã‚Œã¾ã™ï¼Žåˆ†å‰²ã®æ–¹æ³•ã«ã‚‚å¹¾ã¤ã‹ã®ç¨®é¡žãŒã‚り，全ã¦ã®åˆ†é¡žå¯¾è±¡ãŒã¡ã‚‡ã†ã©ä¸€ã¤ã ã‘ã®ã‚¯ãƒ©ã‚¹ã‚¿ã®è¦ç´ ã¨ãªã‚‹å ´åˆ(ãƒãƒ¼ãƒ‰ãªã‚‚ã—ãã¯ï¼Œã‚¯ãƒªã‚¹ãƒ—ãªã‚¯ãƒ©ã‚¹ã‚¿ã¨ã„ã„ã¾ã™)や，逆ã«ä¸€ã¤ã®ã‚¯ãƒ©ã‚¹ã‚¿ãŒè¤‡æ•°ã®ã‚¯ãƒ©ã‚¹ã‚¿ã«åŒæ™‚ã«éƒ¨åˆ†çš„ã«æ‰€å±žã™ã‚‹å ´åˆ(ソフト，ã¾ãŸã¯ï¼Œãƒ•ã‚¡ã‚¸ã‚£ãªã‚¯ãƒ©ã‚¹ã‚¿ã¨ã„ã„ã¾ã™)ãŒã‚ã‚Šã¾ã™ï¼Žã“ã“ã§ã¯å‰è€…ã®ãƒãƒ¼ãƒ‰ãªå ´åˆã®ã‚¯ãƒ©ã‚¹ã‚¿ãƒªãƒ³ã‚°ã«ã¤ã„ã¦è¿°ã¹ã¾ã™ï¼Ž
ソフトãªæŽ¨è«–Markov Logic Networkã®ç´¹ä»‹ - Preferred Networks Research & Development
予約ã—ãŸã‚‚ã®ã®ã‚¤ãƒ³ãƒ•ã‚©ãƒãƒ¼ã‚’手ã«å…¥ã‚Œã‚‰ã‚Œãªã„海野ã§ã™ï¼Ž 人間ã®é«˜åº¦ãªçŸ¥çš„処ç†ã®ä¸€ã¤ãŒã€æŽ¨è«–処ç†ã§ã™ï¼Žä»Šæ—¥ã¯ãã®æŽ¨è«–ã‚’ã€è¿°èªžè«–ç†ã¨æ©Ÿæ¢°å¦ç¿’ã®çµ„ã¿åˆã‚ã›ã§æ¨¡å€£ã—ãŸMarkov Logic Networkã¨ã„ã†æ‰‹æ³•ã¨ã€ãã®OSS実装ã§ã‚ã‚‹Alchemyã®ç´¹ä»‹ã§ã™ï¼Ž é³¥ã¨ã¯ãªã‚“ã§ã™ã‹ï¼Ÿã¨ã„ã†è³ªå•ã«å¯¾ã—ã¦ã©ã†ç”ãˆã‚‹ã§ã—ょã†ã‹ï¼Žå¤§é›‘把ã«ã¯ã€ä»¥ä¸‹ã®ã‚ˆã†ã«è€ƒãˆã‚‹ã§ã—ょã†ï¼Ž é³¥ã¨ã¯ã€ç©ºã‚’飛ã¶å‹•ç‰©ã§ã™ï¼Ž ã“ã®å›žç”ã«å¯¾ã—ã¦ã€ã€Œãƒšãƒ³ã‚®ãƒ³ã¯é£›ã°ãªã„よã€ã¨åè«–ã™ã‚‹äººãŒã„ã‚‹ã‹ã‚‚知れã¾ã›ã‚“． é³¥ã¨ã¯ã€ãã¡ã°ã—ã‚’æŒã£ãŸå‹•ç‰©ã§ã™ï¼Ž ã™ã‚‹ã¨ã€ã€Œã‚«ãƒ¢ãƒŽãƒã‚·ã¯é³¥ã˜ã‚ƒãªã„よã€ã¨è¨€ã‚れるã§ã—ょã†ï¼Žäººé–“ã¯åˆã‚ã¦è¦‹ãŸç”Ÿã物ãŒé³¥ã‹ãã†ã˜ã‚ƒãªã„ã‹åˆ¤æ–ã™ã‚‹ã¨ãã€ã©ã†ã—ã¦ã„ã‚‹ã®ã§ã—ょã†ã‹ï¼Žæ€ã†ã«ã€å°‘æ•°ã®è¦å‰‡ï¼ˆé£›ã¶ã‹ã©ã†ã‹ï¼Žãã¡ã°ã—ã‚’ã‚‚ã¤ã‹ï¼‰ã‹ã‚‰ç·åˆçš„ã«åˆ¤æ–ã—ã¦ã„るよã†ã«æ€ã‚ã‚Œã¾ã™ï¼Žäººé–“ã®æŽ¨è«–ã¨ã„ã†ã®ã¯æ¦‚ã以下ã®ã‚ˆã†ãªç‰¹å¾´ã‚’æŒã£ã¦ã„ã‚‹ã®ã§ã¯ãªã„ã‹ã¨
リリースã€éšœå®³æƒ…å ±ãªã©ã®ã‚µãƒ¼ãƒ“スã®ãŠçŸ¥ã‚‰ã›
最新ã®äººæ°—エントリーã®é…ä¿¡
j次ã®ãƒ–ックマーク
kå‰ã®ãƒ–ックマーク
lã‚ã¨ã§èªã‚€
eコメント一覧を開ã
oページを開ã
Hiroshi Nakagawa presentations
ALIZÉ opensource speaker recognition
ãƒã‚¹ã‚¿ãƒ¼ã‚»ãƒƒã‚·ãƒ§ãƒ³ | IBIS2012
Amazon.com: Optimization for Machine Learning (Neural Information Processing Series): Sra, Suvrit, Nowozin, Sebastian, Wright, Stephen J.: Books
http://case.f7.ems.okayama-u.ac.jp/statedu/hbw2-book/node115.html
Kaggle: Your Machine Learning and Data Science Community
{{#tags}}- {{label}}
{{/tags}}