• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹Á¤º¸°úÇÐȸ ³í¹®Áö > Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)

Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) ±×·¡ÇÁ ±â¹Ý ÁØÁöµµ ÇнÀ¿¡¼­ ºü¸¥ ³·Àº °è¼ö Ç¥Çö ±â¹Ý ±×·¡ÇÁ ±¸Ãà
¿µ¹®Á¦¸ñ(English Title) Graph Construction Based on Fast Low-Rank Representation in Graph-Based Semi-Supervised Learning
ÀúÀÚ(Author) ¿Àº´È­   ¾çÁöÈÆ   Byonghwa Oh   Jihoon Yang  
¿ø¹®¼ö·Ïó(Citation) VOL 45 NO. 01 PP. 0015 ~ 0021 (2018. 01)
Çѱ۳»¿ë
(Korean Abstract)
³·Àº °è¼ö Ç¥Çö(Low-Rank Representation, LRR) ±â¹Ý ¹æ¹ýÀº ¾ó±¼ Ŭ·¯½ºÅ͸µ, °´Ã¼ °ËÃâ µîÀÇ ¿©·¯ ½ÇÁ¦ ÀÀ¿ë¿¡ ³Î¸® »ç¿ëµÇ°í ÀÖ´Ù. ÀÌ ¹æ¹ýÀº ±×·¡ÇÁ ±â¹Ý ÁØÁöµµ ÇнÀ¿¡¼­ ±×·¡ÇÁ ±¸Ãà¿¡ »ç¿ëÇÒ °æ¿ì ³ôÀº ¿¹Ãø Á¤È®µµ¸¦ È®º¸ÇÒ ¼ö ÀÖ¾î ¸¹ÀÌ »ç¿ëµÈ´Ù. ±×·¯³ª LRR ¹®Á¦¸¦ ÇØ°áÇϱâ À§Çؼ­´Â ¾Ë°í¸®ÁòÀÇ ¸Å ¹Ýº¹¸¶´Ù µ¥ÀÌÅÍ ¼ö Å©±âÀÇ Á¤¹æÇà·Ä¿¡ ´ëÇØ Æ¯ÀÌ°ª ºÐÇظ¦ ¼öÇàÇÏ¿©¾ß ÇϹǷΠ°è»ê ºñÈ¿À²ÀûÀÌ´Ù. À̸¦ ÇØ°áÇϱâ À§ÇØ ¼Óµµ¸¦ Çâ»ó½ÃŲ ¹ßÀüµÈ LRR ¹æ¹ýÀ» Á¦¾ÈÇÑ´Ù. ÀÌ´Â ÃÖ±Ù ¹ßÇ¥µÈ Fast LRR(FaLRR)À» ±â¹ÝÀ¸·Î Çϸç, FaLRRÀÌ ¼Óµµ´Â ºü¸£Áö¸¸ ½ÇÁ¦·Î ºÐ·ù ¹®Á¦¿¡¼­ ¼º´ÉÀÌ ³·Àº °ÍÀ» ÇØ°áÇϱâ À§ÇØ ±â¹Ý ÃÖÀûÈ­ ¸ñÇ¥¿¡ Ãß°¡ Á¦¾à Á¶°ÇÀ» µµÀÔÇÏ°í À̸¦ ÃÖÀûÈ­ÇÏ´Â ¹æ¹ýÀ» Á¦¾ÈÇÑ´Ù. ½ÇÇèÀ» ÅëÇÏ¿© Á¦¾È ¹æ¹ýÀº LRRº¸´Ù ´õ ÁÁÀº Çظ¦ ºü¸£°Ô ã¾Æ³¿À» È®ÀÎÇÒ ¼ö ÀÖ´Ù. ¶ÇÇÑ, µ¿ÀÏÇÑ Çظ¦ µµÃâÇÏ´Â ¹æ¹ýÀ» ã¾Æ³»±â´Â ¾î·ÆÁö¸¸ ÃÖ¼ÒÈ­ÇÏ´Â ¸ñÇ¥°¡ Ãß°¡µÉ °æ¿ì ´õ ÁÁÀº °á°ú¸¦ ³ªÅ¸³»´Â Fast MLRR(FaMLRR)À» Á¦¾ÈÇÑ´Ù.
¿µ¹®³»¿ë
(English Abstract)
Low-Rank Representation (LRR) based methods are widely used in many practical applications, such as face clustering and object detection, because they can guarantee high prediction accuracy when used to constructing graphs in graph – based semi-supervised learning. However, in order to solve the LRR problem, it is necessary to perform singular value decomposition on the square matrix of the number of data points for each iteration of the algorithm; hence the calculation is inefficient. To solve this problem, we propose an improved and faster LRR method based on the recently published Fast LRR (FaLRR) and suggests ways to introduce and optimize additional constraints on the underlying optimization goals in order to address the fact that the FaLRR is fast but actually poor in classification problems. Our experiments confirm that the proposed method finds a better solution than LRR does. We also propose Fast MLRR (FaMLRR), which shows better results when the goal of minimizing is added.
Å°¿öµå(Keyword) ÁØÁöµµ ÇнÀ   ±×·¡ÇÁ ±â¹Ý ÁØÁöµµ ÇнÀ   ±×·¡ÇÁ ±¸Ãà   ³·Àº °è¼ö Ç¥Çö   semi-supervised learning   graph-based semi-supervised learning   graph construction   low-rank representation  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå