Neural Networks with Few Multiplications
For most deep learning algorithms training is notoriously time consuming. Since most of the computation in training neural networks is typically spent on floating point multiplications, we investigate an approach to training that eliminates the need for most of these. Our method consists of two parts: First we stochastically binarize weights to convert multiplications involved in computing hidden
Eigen
Get it The latest stable release is Eigen 3.4.0. Get it here: tar.bz2, tar.gz, zip. Changelog. The latest 3.3 release is Eigen 3.3.9. Get it here: tar.bz2, tar.gz, zip. Changelog. The latest 3.2 release is Eigen 3.2.10. Get it here: tar.bz2, tar.gz, zip. Changelog. The unstable source code from the master is there: tar.bz2, tar.gz, zip. To check out the Eigen repository using Git, do: git clone ht
GraphLab: A New Parallel Framework for Machine Learning
GraphLab: A Parallel Framework for Machine LearningDesigning and implementing efficient and provably correct parallel machine learning (ML) algorithms can be very challenging. Existing high-level parallel abstractions like MapReduce are often insufficiently expressive while low-level tools like MPI and Pthreads leave ML experts repeatedly solving the same design challenges. By targeting common pat
ç§ã®ãƒ–ックマーク ： 簡潔データ構é€
田部井é–生（科å¦æŠ€è¡“振興機構 ERATOæ¹Šé›¢æ•£æ§‹é€ å‡¦ç†ç³»ãƒ—ãƒã‚¸ã‚§ã‚¯ãƒˆç ”究員) ã¯ã˜ã‚ã« è¿‘å¹´ã€Web技術や計測技術ã®ç™ºå±•ã«ã‚ˆã‚Šè¨€èªžã‚„ゲノムデータã¯å¤§è¦æ¨¡åŒ–ã—ã¦ã„ã¾ã™ã€‚従æ¥ã®ãƒ‡ãƒ¼ã‚¿æ§‹é€ ã¯å¤§è¦æ¨¡ãƒ‡ãƒ¼ã‚¿ã‚’扱ã†ã«ã¯ã‚µã‚¤ã‚ºãŒå¤§ããメモリã«è¼‰ã‚‰ãªã„〠ã—ã‹ã—ã€åœ§ç¸®ã™ã‚‹ã¨ãƒ©ãƒ³ãƒ€ãƒ アクセスをã™ã‚‹ã“ã¨ãŒã§ããªã„ã¨ã„ã†æ¬ 点ãŒã‚ã‚Šã¾ã™ã€‚ ç°¡æ½”ãƒ‡ãƒ¼ã‚¿æ§‹é€ ã¨ã¯ãƒ‡ãƒ¼ã‚¿ã‚’å°ã•ãä¿å˜ã‹ã¤é«˜é€Ÿãªæ“作ãŒå¯èƒ½ãªãƒ‡ãƒ¼ã‚¿æ§‹é€ ã§ã™ã€‚ è¿‘å¹´ã€é›†åˆã€æ–‡å—列ã€æœ¨ã€ã‚°ãƒ©ãƒ•ãƒ‡ãƒ¼ã‚¿ã‚’扱ã†ãŸã‚ã®ç°¡æ½”ãƒ‡ãƒ¼ã‚¿æ§‹é€ ãŒæ案ã•ã‚Œæ³¨ç›®ã‚’集ã‚ã¦ã„ã¾ã™ã€‚ ç§ãŸã¡ã®èº«è¿‘ãªã‚¢ãƒ—リケーションã¨ã—ã¦ã€Google日本語入力ã§ã¯ç°¡æ½”木LOUDSã®å®Ÿè£…ãŒä½¿ã‚ã‚Œã€å®Ÿéš›ã«ä½¿ã‚ã‚Œã¯ã˜ã‚ã¦ã„ã¾ã™ã€‚ ã¾ãŸã€æœ‰å¿—ã«ã‚ˆã‚‹ãれらを解説ã—ãŸã‚µã‚¤ãƒˆã‚„ライブラリãªã©ã‚‚利用å¯èƒ½ã«ãªã‚Šã¤ã¤ã‚ã‚Šã¾ã™ã€‚ ãã“ã§ã€ã“ã®ãƒšãƒ¼ã‚¸ã§ã¯ç°¡æ½”ãƒ‡ãƒ¼ã‚¿æ§‹é€ ã‚’ç”¨ã„ãŸç ”究開発ã®ãŸã‚ã®ã„ã‚ã„ã‚ãªãƒªã‚½ãƒ¼ã‚¹ã‚’紹介ã—ã¾ã™ã€‚ 解説記
1
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¿¡
処ç†ã‚’実行ä¸ã§ã™
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
{{#tags}}- {{label}}
{{/tags}}