{
"title": "An Evolutionary Statistical Learning Theory",
"authors": "Sung-Hae Jun, Kyung-Whan Oh",
"country": "Korea, Republic Of",
"institution": "Cheongju University",
"volume": "12",
"journal": "International Journal of Computer, Electrical, Automation, Control and Information Engineering",
"pagesStart": 3873,
"pagesEnd": 3881,
"ISSN": "1307-6892",
"URL": "http:\/\/waset.org\/publications\/5739",
"abstract": "Statistical learning theory was developed by Vapnik. It\r\nis a learning theory based on Vapnik-Chervonenkis dimension. It also\r\nhas been used in learning models as good analytical tools. In general, a\r\nlearning theory has had several problems. Some of them are local\r\noptima and over-fitting problems. As well, statistical learning theory\r\nhas same problems because the kernel type, kernel parameters, and\r\nregularization constant C are determined subjectively by the art of\r\nresearchers. So, we propose an evolutionary statistical learning theory\r\nto settle the problems of original statistical learning theory.\r\nCombining evolutionary computing into statistical learning theory,\r\nour theory is constructed. We verify improved performances of an\r\nevolutionary statistical learning theory using data sets from KDD cup.",
"references": null,
"publisher": "World Academy of Science, Engineering and Technology",
"index": "International Science Index 12, 2007"
}