\input zb-basic
\input zb-ioport
\iteman{io-port 05989192}
\itemau{Feng, Yun-Long; Lv, Shao-Gao}
\itemti{Unified approach to coefficient-based regularized regression.}
\itemso{Comput. Math. Appl. 62, No. 1, 506-515 (2011).}
\itemab
Summary: We consider the coefficient-based regularized least-squares regression problem with the $l^{q}$-regularizer ($1\le q\le 2$) and data dependent hypothesis spaces. Algorithms for data dependent hypothesis spaces perform well with the property of flexibility. We conduct a unified error analysis by a stepping stone technique. An empirical covering number technique is also employed in our study to improve sample errors. Comparing with existing results, we make a few improvements: First, we obtain a significantly sharper learning rate that can be arbitrarily close to $O(m^{ - 1})$ under reasonable conditions, which is regarded as the best learning rate in learning theory. Second, our results cover the case $q=1$, which is novel. Finally, our results hold under very general conditions.
\itemrv{~}
\itemcc{}
\itemut{data dependent hypothesis spaces; $l^{q}$-regularizer; $\ell ^{2}$-empirical covering number; learning rates}
\itemli{doi:10.1016/j.camwa.2011.05.034}
\end