Share Email Print

Proceedings Paper

Pseudo orthogonal bases give the optimal generalization capability in neural network learning
Author(s): Masashi Sugiyama; Hidemitsu Ogawa
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Pseudo orthogonal bases are a certain type of frames proposed in the engineering field, whose concept is equivalent to a tight frame with frame bound 1 in the frame terminology. This paper shows that pseudo orthogonal bases play an essential role in neural network learning. One of the most important issues in neural network learning is `what training data provides the optimal generalization capability?', which is referred to as active learning in the neural network community. We derive a necessary and sufficient condition of training data to provide the optimal generalization capability in the trigonometric polynomial space, where the concept of pseudo orthogonal bases is essential. By utilizing useful properties of pseudo orthogonal bases, we clarify the mechanism of achieving the optimal generalization.

Paper Details

Date Published: 26 October 1999
PDF: 12 pages
Proc. SPIE 3813, Wavelet Applications in Signal and Image Processing VII, (26 October 1999); doi: 10.1117/12.366809
Show Author Affiliations
Masashi Sugiyama, Tokyo Institute of Technology (Japan)
Hidemitsu Ogawa, Tokyo Institute of Technology (Japan)

Published in SPIE Proceedings Vol. 3813:
Wavelet Applications in Signal and Image Processing VII
Michael A. Unser; Akram Aldroubi; Andrew F. Laine, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?