• 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¸Ã³¸®ÇÐȸ ³í¹®Áö > Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ

Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ

Current Result Document : 249 / 249

ÇѱÛÁ¦¸ñ(Korean Title) ¿µ»ó µ¥ÀÌÅÍ Æ¯Â¡ Ä¿¹ö¸®Áö ±â¹Ý µö·¯´× ¸ðµ¨ °ËÁõ ±â¹ý
¿µ¹®Á¦¸ñ(English Title) Deep Learning Model Validation Method Based on Image Data Feature Coverage
ÀúÀÚ(Author) ÀÓâ³²   ¹Ú¿¹½½   ÀÌÁ¤¿ø   Chang-Nam Lim   Ye-Seul Park   Jung-Won Lee  
¿ø¹®¼ö·Ïó(Citation) VOL 10 NO. 09 PP. 0375 ~ 0384 (2021. 09)
Çѱ۳»¿ë
(Korean Abstract)
µö·¯´× ±â¹ýÀº ¿µ»ó ó¸® ºÐ¾ß¿¡¼­ ³ôÀº ¼º´ÉÀ» ÀÔÁõ ¹Þ¾Æ ´Ù¾çÇÑ ºÐ¾ß¿¡¼­ Àû¿ëµÇ°í ÀÖ´Ù. ÀÌ·¯ÇÑ µö·¯´× ¸ðµ¨ÀÇ °ËÁõ¿¡ °¡Àå ³Î¸® »ç¿ëµÇ´Â ¹æ¹ýÀ¸·Î´Â Ȧµå¾Æ¿ô °ËÁõ ¹æ¹ý, k-°ã ±³Â÷ °ËÁõ ¹æ¹ý, ºÎÆ®½ºÆ®·¦ ¹æ¹ý µîÀÌ ÀÖ´Ù. ÀÌ·¯ÇÑ ±âÁ¸ÀÇ ±â¹ýµéÀº µ¥ÀÌÅÍ ¼ÂÀ» ºÐÇÒÇÏ´Â °úÁ¤¿¡¼­ Ŭ·¡½º °£ÀÇ ºñÀ²¿¡ ´ëÇÑ ±ÕÇüÀ» °í·ÁÇÏÁö¸¸, °°Àº Ŭ·¡½º ³»¿¡¼­µµ Á¸ÀçÇÏ´Â ´Ù¾çÇÑ Æ¯Â¡µéÀÇ ºñÀ²Àº °í·ÁÇÏÁö ¾Ê°í ÀÖ´Ù. ÀÌ·¯ÇÑ Æ¯Â¡µéÀ» °í·ÁÇÏÁö ¾ÊÀ» °æ¿ì, ÀϺΠƯ¡¿¡ ÆíÇâµÈ °ËÁõ °á°ú¸¦ ¾ò°Ô µÉ ¼ö ÀÖ´Ù. µû¶ó¼­ º» ³í¹®¿¡¼­´Â ±âÁ¸ °ËÁõ ¹æ¹ýµéÀ» °³¼±ÇÏ¿© ¿µ»ó ºÐ·ù¸¦ À§ÇÑ µ¥ÀÌÅÍ Æ¯Â¡ Ä¿¹ö¸®Áö ±â¹ÝÀÇ µö·¯´× ¸ðµ¨ °ËÁõ ±â¹ýÀ» Á¦¾ÈÇÑ´Ù. Á¦¾ÈÇÏ´Â ±â¹ýÀº µö·¯´× ¸ðµ¨ÀÇ ÇнÀ°ú °ËÁõÀ» À§ÇÑ ÈÆ·Ã µ¥ÀÌÅÍ ¼Â°ú Æò°¡ µ¥ÀÌÅÍ ¼ÂÀÌ Àüü µ¥ÀÌÅÍ ¼ÂÀÇ Æ¯Â¡À» ¾ó¸¶³ª ¹Ý¿µÇÏ°í ÀÖ´ÂÁö ¼öÄ¡·Î ÃøÁ¤ÇÒ ¼ö ÀÖ´Â µ¥ÀÌÅÍ Æ¯Â¡ Ä¿¹ö¸®Áö¸¦ Á¦¾ÈÇÑ´Ù. ÀÌ·¯ÇÑ ¹æ½ÄÀº Àüü µ¥ÀÌÅÍ ¼ÂÀÇ Æ¯Â¡À» ¸ðµÎ Æ÷ÇÔÇϵµ·Ï Ä¿¹ö¸®Áö¸¦ º¸ÀåÇÏ¿© µ¥ÀÌÅÍ ¼ÂÀ» ºÐÇÒÇÒ ¼ö ÀÖ°í, ¸ðµ¨ÀÇ Æò°¡ °á°ú¸¦ »ý¼ºÇÑ Æ¯Â¡ ±ºÁý ´ÜÀ§·Î ºÐ¼®ÇÒ ¼ö ÀÖ´Ù. °ËÁõ °á°ú, ÈÆ·Ã µ¥ÀÌÅÍ ¼ÂÀÇ µ¥ÀÌÅÍ Æ¯Â¡ Ä¿¹ö¸®Áö°¡ ³·¾ÆÁú °æ¿ì, ¸ðµ¨ÀÌ Æ¯Á¤ Ư¡¿¡ ÆíÇâµÇ°Ô ÇнÀÇÏ¿© ¸ðµ¨ÀÇ ¼º´ÉÀÌ ³·¾ÆÁö¸ç, Fashion-MNISTÀÇ °æ¿ì Á¤È®µµ°¡ 8.9%±îÁö Â÷À̳ª´Â °ÍÀ» È®ÀÎÇÏ¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
Deep learning techniques have been proven to have high performance in image processing and are applied in various fields. The most widely used methods for validating a deep learning model include a holdout verification method, a k-fold cross verification method, and a bootstrap method. These legacy methods consider the balance of the ratio between classes in the process of dividing the data set, but do not consider the ratio of various features that exist within the same class. If these features are not considered, verification results may be biased toward some features. Therefore, we propose a deep learning model validation method based on data feature coverage for image classification by improving the legacy methods. The proposed technique proposes a data feature coverage that can be measured numerically how much the training data set for training and validation of the deep learning model and the evaluation data set reflects the features of the entire data set. In this method, the data set can be divided by ensuring coverage to include all features of the entire data set, and the evaluation result of the model can be analyzed in units of feature clusters. As a result, by providing feature cluster information for the evaluation result of the trained model, feature information of data that affects the trained model can be provided.
Å°¿öµå(Keyword) µö·¯´×   ¸ðµ¨ Å×½ºÆà  ¿µ»ó Ư¡ ÃßÃâ   °ËÁõ ±â¹ý   µ¥ÀÌÅÍ ¼Â ºÐÇÒ ±â¹ý   Deep Learning   Coverage Testing   Image Feature Extraction   Validation Method   Dataset Splitting Method  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå