Á¤º¸°úÇÐȸ ÄÄÇ»ÆÃÀÇ ½ÇÁ¦ ³í¹®Áö (KIISE Transactions on Computing Practices)
ÇѱÛÁ¦¸ñ(Korean Title) |
GPGPU¸¦ È°¿ëÇÑ Àΰø½Å°æ¸Á ¿¹Ãø±â¹Ý ÅؽºÆ® ¾ÐÃà±â¹ý |
¿µ¹®Á¦¸ñ(English Title) |
Neural Predictive Coding for Text Compression Using GPGPU |
ÀúÀÚ(Author) |
±èÀçÁÖ
ÇÑȯ¼ö
Jaeju Kim
Hwansoo Han
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 22 NO. 03 PP. 0127 ~ 0132 (2016. 03) |
Çѱ۳»¿ë (Korean Abstract) |
Àΰø½Å°æ¸ÁÀ» ¾ÐÃà¿¡ Àû¿ëÇÏ¿© ´õ ³ôÀº ¾ÐÃà ¼º´ÉÀ» º¸À̱â À§ÇÑ ¾Ë°í¸®ÁòµéÀÌ ¸î °¡Áö ¿¬±¸µÇ¾î ÀÖ´Ù. ±×·¯³ª ±×µ¿¾È ÀÌ·¯ÇÑ ¾Ë°í¸®ÁòµéÀº ÇÑÁ¤µÈ °è»ê ´É·ÂÀÇ Çϵå¿þ¾î¸¦ °¡Áö°í Àֱ⿡ ÀÛÀº Å©±âÀÇ ½Å°æ¸ÁÀ» »ç¿ëÇÒ ¼ö¹Û¿¡ ¾ø¾úÀ¸¸ç Àû¿ëÇÏ´Â ´ë»ó ¿ª½Ã ½ÇÁ¦·Î »ç¿ëÇϱ⿡´Â ³Ê¹« ÀÛÀº Å©±âÀÇ ÆÄÀϵéÀ̾ú´Ù. º» ³í¹®¿¡¼´Â GPGPUÀÇ °è»ê´É·ÂÀ» ½Å°æ¸Á ÇнÀ¿¡ ÀÌ¿ëÇÏ¿© ¸¸µç ÅؽºÆ® ¹®¸Æ ±â¹Ý ¹®ÀÚ µîÀå È®·ü ¿¹Ãø±â¿Í ÇÔ²² ÇãÇÁ¸¸ ºÎÈ£ÈÀÇ ¼º´ÉÀ» ³ôÀÏ ¼ö ÀÖ´Â º¯È¯ ¹æ¹ýÀ» Á¦½ÃÇÑ´Ù. ¾Õ¸ÔÀÓ ½Å°æ¸Á°ú GRU ȸ±Í ½Å°æ¸Á¿¡ ´ëÇØ ½ÇÇèÀ» ¼öÇàÇÏ¿´À¸¸ç, ȸ±Í ½Å°æ¸Á ¸ðµ¨Àº ¾Õ¸ÔÀÓ ½Å°æ¸Á¿¡ ºñÇØ ¶Ù¾î³ ¿¹Ãø ¼º°ø·ü°ú ¾ÐÃà·üÀ» º¸¿´´Ù.
|
¿µ¹®³»¿ë (English Abstract) |
Several methods have been proposed to apply artificial neural networks to text compression in the past. However, the networks and targets are both limited to the small size due to hardware capability in the past. Modern GPUs have much better calculation capability than CPUs in an order of magnitude now, even though CPUs have become faster. It becomes possible now to train greater and complex neural networks in a shorter time. This paper proposed a method to transform the distribution of original data with a probabilistic neural predictor. Experiments were performed on a feedforward neural network and a recurrent neural network with gated-recurrent units. The recurrent neural network model outperformed feedforward network in compression rate and prediction accuracy.
|
Å°¿öµå(Keyword) |
Àΰø½Å°æ¸Á
¾Õ¸ÔÀÓ ½Å°æ¸Á
ȸ±Í½Å°æ¸Á
ÅؽºÆ® ¾ÐÃà
ÀÚ¿¬¾î ¾ÐÃà
ÇãÇÁ¸¸ ºÎÈ£È
¿£Æ®
artificial neural network
feedforward neural network
recurrent neural network
text compression
natural language compression
huffman coding
entropy coding
batch normalization
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|