Title :
Complexity regularization using data-dependent penalties
Author :
Lugosi, GBbor ; Nobel, Andrew
Author_Institution :
Dept. of Math., Tech. Univ. Budapest, Hungary
Abstract :
We define a regression function estimate based on complexity regularization, where the list of candidate functions and the corresponding penalties are determined from the training data, leading to improved performance
Keywords :
statistical analysis; candidate functions; complexity regularization; data-dependent penalties; regression function estimate; training data; Artificial intelligence; Bismuth; Mathematics; Random variables; Statistics; Training data;
Conference_Titel :
Information Theory, 1995. Proceedings., 1995 IEEE International Symposium on
Conference_Location :
Whistler, BC
Print_ISBN :
0-7803-2453-6
DOI :
10.1109/ISIT.1995.535769