See this document in CiteSeerX!

Subspace Information Criterion for Sparse Regressors (2001)  (Make Corrections)  
Koji Tsuda, Masashi Sugiyama, Klaus-Robert Müller



  Home/Search   Context   Related

 
View or download:
ida.first.fhg.de/pu...TsuSugMue01.ps.gz
Cached:  PS.gz  PS  PDF   Image  Update  Help

From:  ida.first.fraunhof...publications (more)
(Enter author homepages)

Rate this article: (best)
  Comment on this article  
(Enter summary)

Abstract: Non-quadratic regularizers, in particular the ` 1 norm regularizer can yield sparse solutions that generalize well. In this work we propose the Generalized Subspace Information Criterion (GSIC) that allows to predict the generalization error for this useful family of regularizers. We show that under some technical assumptions GSIC is an asymptotically unbiased estimator of the generalization error. GSIC is demonstrated to have a good performance in experiments with the ` 1 norm regularizer ... (Update)

Similar documents (at the sentence level):
75.3%:   Subspace Information Criterion for Sparse Regressors - Tsuda, Sugiyama, Müller (2001)   (Correct)
47.3%:   Subspace Information Criterion for Non-Quadratic.. - Tsuda, Sugiyama, Müller (2002)   (Correct)

Active bibliography (related documents):   More   All
0.2:   Improving Precision of the Subspace Information Criterion - Sugiyama (2003)   (Correct)
0.2:   Theoretical and Experimental Evaluation of Subspace.. - Sugiyama, Ogawa (2001)   (Correct)
0.2:   The Subspace Information Criterion for Infinite Dimensional.. - Sugiyama, al. (2002)   (Correct)

BibTeX entry:   (Update)

@misc{ tsuda-subspace,
  author = "Koji Tsuda and Masashi Sugiyama and Klaus-Robert Müller",
  title = "Subspace Information Criterion for Sparse Regressors",
  url = "citeseer.ist.psu.edu/tsuda01subspace.html" }
Citations (may not include all citations):
947   Statistical Learning Theory (context) - Vapnik - 1998
244   Bayesian interpolation - MacKay - 1992
132   Theory of Optimal Experiments (context) - Fedorov - 1972
124   Robust linear programming discrimination of two linearly ins.. - Bennett, Mangasarian - 1992
107   Some comments on CP (context) - Mallows - 1973
85   Network information criterion --- determining the number of .. - Murata, Yoshizawa et al. - 1994
30   Introduction to radial basis function networks (context) - Orr - 1996
22   Subspace information criterion for model selection - Sugiyama, Ogawa - 2001
21   Mathematical programming in data mining - Mangasarian - 1997
21   A mathematical programming approach to the kernel fisher alg.. - Mika, Ratsch et al. - 2001
13   Theoretical and experimental evaluation of subspace informat.. - Sugiyama, Ogawa - 2001
10   Bayesian regularisation and pruning using a Laplace prior - Williams - 1995
8   SIAM: Society for Industrial and Applied Mathematics (context) - Wahba, for et al. - 1990
8   Subspace information criterion for non-quadratic regularizer.. - Tsuda, Sugiyama et al. - 2000

Documents on the same site (http://ida.first.fraunhofer.de/homepages/ida/publications.html):   More
Optimizing Property Codes in Protein Data Reveals.. - Weiss, Ziehe, Herzel   (Correct)
On-line Learning in Changing Environments with.. - Murata, Kawanabe, .. (2002)   (Correct)
The Leave-one-out Kernel - Tsuda, Kawanabe   (Correct)

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC