Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)
Current Result Document : 1 / 1
ÇѱÛÁ¦¸ñ(Korean Title) |
´Ü¼øÈÇÑ ÇÁ·¹¼Â °Å¸®¸¦ ÀÌ¿ëÇÑ Àû´ëÀû »ý¼º ½Å°æ¸ÁÀÇ ¸ðµå µå·Ó ¹× ºØ±« °ËÃâ ±â¹ý |
¿µ¹®Á¦¸ñ(English Title) |
Detecting Mode Drop and Collapse in GANs Using Simplified Frèchet Distance |
ÀúÀÚ(Author) |
±èÃæÀÏ
Á¤½Â¿ø
¹®ÁöÈÆ
ȲÀÎÁØ
Chung-Il Kim
Seungwon Jung
Jihoon Moon
Eenjun Hwang
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 46 NO. 10 PP. 1012 ~ 1019 (2019. 10) |
Çѱ۳»¿ë (Korean Abstract) |
Àû´ëÀû »ý¼º ½Å°æ¸ÁÀº µÎ °³ÀÇ ³×Æ®¿öÅ©¸¦ Àû´ëÀûÀ¸·Î ÇнÀ½ÃÄÑ ¿øº» µ¥ÀÌÅÍ ºÐÆ÷¸¦ ÃßÁ¤ÇÏ°í, À̸¦ ±â¹ÝÀ¸·Î µ¥ÀÌÅ͸¦ »ý¼ºÇϴµ¥ Ź¿ùÇÑ ¸ðµ¨ÀÌÁö¸¸, ÇнÀ µµÁß ºÐÆ÷¸¦ ÇнÀÇÏÁö ¸øÇÏ´Â ¸ðµå µå·Ó Çö»óÀ̳ª Çϳª ¶Ç´Â ¸Å¿ì ÀûÀº ºÐÆ÷ÀÇ »ùÇø¸À» »ý¼ºÇÏ´Â ¸ðµå ºØ±« Çö»óÀÌ Á¾Á¾ ³ªÅ¸³´Ù. ÀÌ Çö»óÀ» °¨ÁöÇϱâ À§ÇØ ±âÁ¸ ¿¬±¸µéÀº ÇнÀ µ¥ÀÌÅ͸¦ ÅëÁ¦Çϰųª º°µµÀÇ ½Å°æ¸Á ¸ðµ¨À» ÇнÀ½ÃÄÑ¾ß ÇÏ´Â ÇÑ°èÁ¡À» º¸¿´´Ù. ÀÌ¿¡ º» ³í¹®Àº ÇÁ·¹¼Â °Å¸®¸¦ ´Ü¼øÈÇÏ¿© Ãß°¡ÀûÀÎ ¸ðµ¨À̳ª ÇнÀ µ¥ÀÌÅÍÀÇ Á¦ÇÑ ¾øÀÌ ¸ðµå ºØ±«¸¦ °ËÃâÇÏ´Â ±â¹ýÀ» Á¦¾ÈÇÑ´Ù. ´Ù¾çÇÑ ½ÇÇèÀ» ÅëÇÏ¿© Á¦¾ÈÇÏ´Â °Å¸® ôµµ°¡ ±âÁ¸ÀÇ Àû´ëÀû »ý¼º ½Å°æ¸Á¿¡ Àû¿ëµÈ ôµµ¿¡ ºñÇØ ´õ¿í È¿°úÀûÀ¸·Î ¸ðµå µå·Ó ¹× ºØ±«¸¦ °ËÃâÇÒ ¼ö ÀÖÀ½À» º¸ÀδÙ.
|
¿µ¹®³»¿ë (English Abstract) |
Even though generative adversarial network (GAN) is an excellent model for generating data based on the estimation of real data distribution by of two adversarial learning network, this model often suffers from mode drop that does not learn distribution during learning, or mode collapse that generates only one or very few distribution samples. Most studies to detect these problems have used well-balanced data or additional neural network models. In this paper, we propose a method to detect mode drop and collapse by using a simplified Frèchet distance, which does not require any additional model or well-balanced data. Through various experiments, we showed that our proposed distance metric detected mode drop and collapse more accurately than any other metrics used in GANs.
|
Å°¿öµå(Keyword) |
Àû´ëÀû »ý¼º ½Å°æ¸Á
¸ðµå µå·Ó
¸ðµå ºØ±«
°Å¸® Ãøµµ
Generative adversarial nets
mode drop
mode collapse
distance measure
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|