[B! kernel] yu4uã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

yu4u id:yu4u

kernelã«é–¢ã™ã‚‹yu4uã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (5)

${{author_name}}$

{{{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}}

æ–‡å—åˆ—ã‚«ãƒ¼ãƒãƒ«ã«ã‚ˆã‚‹è¾žæ›¸ãªã—ãƒ„ã‚¤ãƒ¼ãƒˆåˆ†é¡ž ã€œæ–‡å—åˆ—ã‚«ãƒ¼ãƒãƒ«å…¥é–€ã€œ
ã‚¼ãƒã‹ã‚‰å§‹ã‚ã‚‹æ·±å±¤å¼·åŒ–å¦ç¿’ï¼ˆNLP2018è¬›æ¼”è³‡æ–™ï¼‰/ Introduction of Deep Reinforcement Learning
yu4u 2011/09/13
string

kernel

twitter

nlp
ãƒªãƒ³ã‚¯
ãƒªãƒ³ã‚¯è§£æž: von Neumann ã‚«ãƒ¼ãƒãƒ« vs. æ£å‰‡åŒ–ãƒ©ãƒ—ãƒ©ã‚·ã‚¢ãƒ³ã¨ã‹ - smlyâ€™s notepad
è¬›ç¾©è³‡æ–™èªã¿+å®¿é¡Œã‚’çµ‚ã‚‰ã›ãŸ. ä»Šå¾Œã‚‚ã£ã¨æœ¬ã‚„ã‚‰è«–æ–‡ã‚„ã‚‰ã‚’èªã¿ã¾ãã£ã¦ä¿®å£«ã§ã®ç ”ç©¶ãƒ†ãƒ¼ãƒžã‚’æ—©ã‚ã«æ˜Žç¢ºã«ã—ã¦ã„ããŸã„. å‰å›žã«ç¶šã„ã¦ã¾ãŸã€Œãƒªãƒ³ã‚¯è§£æžã¨ãã®å‘¨è¾ºã®è©±é¡Œã€ã‚’èªã‚“ã è¨˜éŒ²(ä¸‰æ—¥, å››æ—¥ç›®)ã§ã™ãŒ, å†…å®¹ã«ã¤ã„ã¦ã¯ã‚¹ãƒ©ã‚¤ãƒ‰ã‚’èªã‚ã°ã‚ã‹ã‚‹ã®ã§, ãã®è£œè¶³çš„ãªã“ã¨ã®ã¿ã‚’è¨˜è¿°ã—ã¾ã™. ã—ã‹ã—èª¿åã«ä¹—ã£ã¦å®šç†ã®è¨¼æ˜Žã‚’ã—ã‚ˆã†ã¨ã—ã¦ã†ã¾ãã„ã‹ãªãã¦æ‚¶çµ¶. von Neumann ã‚«ãƒ¼ãƒãƒ«ã¨æ£å‰‡åŒ–ãƒ©ãƒ—ãƒ©ã‚·ã‚¢ãƒ³ã®æ¯”è¼ƒãŒã§ããŸã®ã§ã‚ˆã‹ã£ãŸã§ã™ãŒ, ãã‚Œãã‚‰ã„ã—ã‹å†…å®¹ã‚ã‚Šã¾ã›ã‚“:(ãªãŠãƒ—ãƒã‚°ãƒ©ãƒ ã¯ gists ã®åŸ‹ã‚è¾¼ã¿ãªã®ã§ LDR ã‚„ fastladder ã‹ã‚‰ã¯èªã‚ã¾ã›ã‚“.*1 è¨¼æ˜Žã‚„ãƒ—ãƒã‚°ãƒ©ãƒ ã«èª¤ã‚Šãªã©ãŒã‚ã‚Œã°æŒ‡æ‘˜ã—ã¦ã„ãŸã ã‘ã‚‹ã¨åŠ©ã‹ã‚Šã¾ã™. ãŠã—ãªãŒãã‚¹ãƒ©ã‚¤ãƒ‰ã®è¨‚æ£: ä¸‰æ—¥ç›® 54ãƒšãƒ¼ã‚¸ã®æŒ‡æ•°æ‹¡æ•£ã‚«ãƒ¼ãƒãƒ«ã‚’å±•é–‹ã—ãŸã¨ã“ã‚ von Neumann kernel ã«ã‚ˆã‚‹é‡è¦åº¦ã¨æ£å‰‡åŒ–ãƒ©ãƒ—ãƒ©ã‚·ã‚¢ãƒ³ã«ã‚ˆ
yu4u 2011/06/21
graph laplacian

kernel

graph
ãƒªãƒ³ã‚¯
Fast Subtree Kernels on Graphs
In this article, we propose fast subtree kernels on graphs. On graphs with n nodes and m edges and maximum degree d, these kernels comparing subtrees of height h can be computed in O(mh), whereas th Read more
yu4u 2011/05/16
graph

kernel

ã‚ã¨ã§èªã‚€
ãƒªãƒ³ã‚¯
SDM2011ã§waveletæœ¨ã‚’ç”¨ã„ãŸå¤§è¦æ¨¡ã‚°ãƒ©ãƒ•ãƒ‡ãƒ¼ã‚¿ãƒ™ãƒ¼ã‚¹ã®é«˜é€Ÿé¡žä¼¼åº¦æ¤œç´¢æ‰‹æ³•ã«ã¤ã„ã¦ç™ºè¡¨ã—ã¾ã—ãŸ - Yasuo Tabeiã®æ—¥è¨˜
4æœˆ28æ—¥ã‹ã‚‰4æœˆ30æ—¥ã«é–‹å‚¬ã•ã‚ŒãŸãƒ‡ãƒ¼ã‚¿ãƒ¼ãƒžã‚¤ãƒ‹ãƒ³ã‚°ã®å›½éš›ä¼šè° SIAM Conference on Data Mining (SDM2011)ã«ã¦waveletæœ¨ã‚’ç”¨ã„ãŸå¤§è¦æ¨¡ã‚°ãƒ©ãƒ•ãƒ‡ãƒ¼ã‚¿ãƒ™ãƒ¼ã‚¹ã®é«˜é€Ÿé¡žä¼¼åº¦æ¤œç´¢æ‰‹æ³•ã«ã¤ã„ã¦ç™ºè¡¨ã—ã¦ãã¾ã—ãŸã€‚ Yasuo Tabei and Koji Tsuda: Kernel-based Similarity Search in Massive Graph Databases with Wavelet Trees, Eleventh SIAM International Conference on Data Mining (SDM), Arizona, USA, 2011. Link to the paper æœ¬ç ”ç©¶ã§ã¯å¤§è¦æ¨¡ã‚°ãƒ©ãƒ•ãƒ‡ãƒ¼ã‚¿ãƒ¼ã®ç´¢å¼•ã«ã‚ˆã‚‹é«˜é€Ÿãªé¡žä¼¼åº¦æ¤œç´¢æ‰‹æ³•ã‚’ææ¡ˆã—ã¾ã—ãŸã€‚è¿‘å¹´ã€ã‚°ãƒ©ãƒ•ãƒ‡ãƒ¼ã‚¿ãƒ™ãƒ¼ã‚¹ã«ç™»éŒ²ã•ã‚Œã¦ã„ã‚‹ã‚°ãƒ©ãƒ•ã®æ•°ã¯å¤§è¦æ¨¡åŒ–ã—ã¦ã„
yu4u 2011/05/09
graph

kernel

nns
ãƒªãƒ³ã‚¯
ã‚«ãƒ¼ãƒãƒ«æ³• æ£å®šå€¤ã‚«ãƒ¼ãƒãƒ«ã‚’ç”¨ã„ãŸãƒ‡ãƒ¼ã‚¿è§£æž
1 2004 11 24~26 Final version. Nov.26, 2004 2 I 1. 2. ï¿½ ï¿½ 3. ï¿½ ï¿½ PCA CCA . ï¿½ ï¿½ Bochner ï¿½ representer 3 II 5. ï¿½ ï¿½ ï¿½ ICA, 7. 4 g(x) Parzen window âˆ‘ = âˆ’ = N i i x x g N x p 1 ) ( 1 ) ( 5 1. ï¿½ 6 ï¿½ âŽŸ âŽŸ âŽŸ âŽŸ âŽŸ âŽ âŽž âŽœ âŽœ âŽœ âŽœ âŽœ âŽ âŽ› = N m N m m X X X X X X X L M M L L 1 2 2 1 1 1 1 m N 7 x1 x2 z1 z3 z2 ) 2 , , ( ) , , ( 2 1 2 2 2 1 3 2 1 x x x x z z z = 8 ï¿½ xi Î¦(xi) zi H â„¦ H â„¦ H â†’ â„¦ Î¦ : 9 ï¿½ H = feature space
yu4u 2011/01/20
kernel

tutorial
ãƒªãƒ³ã‚¯
1