Model Selection Paper Available !

[SUGIYAMA Masashi]

Dear colleagues,

I am pleased to announce the availability of our paper on line.

      "Subspace Information Criterion for Model Selection"
            By Masashi Sugiyama and Hidemitsu Ogawa
                To appear in Neural Computation
   http://ogawa-www.cs.titech.ac.jp/~sugi/publications/sic.ps.gz

Also we will give a talk about the above Subspace Information Criterion at

                         NIPS*2000 Workshop
         "Cross-Validation, Bootstrap and Model Selection"
           Breckenridge, Colorado, USA, December 1, 2000.
              http://www.cs.cmu.edu/~rahuls/nips2000/

We appreciate your questions and comments by e-mail or at the workshop.


                             ABSTRACT

The problem of model selection is considerably important for acquiring
higher levels of generalization capability in supervised learning.
In this paper, we propose a new criterion for model selection called
the subspace information criterion (SIC), which is a generalization of
Mallows' $C_L$. It is assumed that the learning target function belongs
to a specified functional Hilbert space and the generalization error is
defined as the Hilbert space squared norm of the difference between the
learning result function and target function. SIC gives an unbiased
estimate of the generalization error so defined. SIC assumes the
availability of an unbiased estimate of the target function and the
noise covariance matrix, which are generally unknown. A practical
calculation method of SIC for least mean squares learning is provided
under the assumption that the dimension of the Hilbert space is less
than the number of training examples. Finally, computer simulations in
two examples show that SIC works well even when the number of training
examples is small. 


Sincerely yours,
Masashi Sugiyama.

---------------------------------
Masashi Sugiyama

Department of Computer Science,
Graduate School of Information Science and Engineering,
Tokyo Institute of Technology,
2-12-1, O-okayama, Meguro-ku, Tokyo, 152-8552, Japan.

E-mail: sugi at og.cs.titech.ac.jp
URL:  http://ogawa-www.cs.titech.ac.jp/~sugi
Tel:  +81-3-5734-2190
Fax:  +81-3-5734-2949