The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Dans cet article, nous considérons le problème de l'apprentissage actif et donnons une condition nécessaire et suffisante de points d'échantillonnage pour une capacité de généralisation optimale. En utilisant les propriétés des bases pseudo orthogonales, nous clarifions le mécanisme permettant d'atteindre la capacité de généralisation optimale. Nous montrons également que la condition fournit non seulement la capacité de généralisation optimale, mais réduit également la complexité de calcul et la mémoire requise pour calculer les fonctions de résultat d'apprentissage. Sur la base de la condition d'optimalité, nous donnons des méthodes de conception de points d'échantillonnage optimaux pour les modèles polynomiaux trigonométriques. Enfin, l'efficacité de la méthode d'apprentissage actif proposée est démontrée par des simulations informatiques.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Masashi SUGIYAMA, Hidemitsu OGAWA, "Active Learning for Optimal Generalization in Trigonometric Polynomial Models" in IEICE TRANSACTIONS on Fundamentals,
vol. E84-A, no. 9, pp. 2319-2329, September 2001, doi: .
Abstract: In this paper, we consider the problem of active learning, and give a necessary and sufficient condition of sample points for the optimal generalization capability. By utilizing the properties of pseudo orthogonal bases, we clarify the mechanism of achieving the optimal generalization capability. We also show that the condition does not only provide the optimal generalization capability but also reduces the computational complexity and memory required to calculate learning result functions. Based on the optimality condition, we give design methods of optimal sample points for trigonometric polynomial models. Finally, the effectiveness of the proposed active learning method is demonstrated through computer simulations.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/e84-a_9_2319/_p
Copier
@ARTICLE{e84-a_9_2319,
author={Masashi SUGIYAMA, Hidemitsu OGAWA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Active Learning for Optimal Generalization in Trigonometric Polynomial Models},
year={2001},
volume={E84-A},
number={9},
pages={2319-2329},
abstract={In this paper, we consider the problem of active learning, and give a necessary and sufficient condition of sample points for the optimal generalization capability. By utilizing the properties of pseudo orthogonal bases, we clarify the mechanism of achieving the optimal generalization capability. We also show that the condition does not only provide the optimal generalization capability but also reduces the computational complexity and memory required to calculate learning result functions. Based on the optimality condition, we give design methods of optimal sample points for trigonometric polynomial models. Finally, the effectiveness of the proposed active learning method is demonstrated through computer simulations.},
keywords={},
doi={},
ISSN={},
month={September},}
Copier
TY - JOUR
TI - Active Learning for Optimal Generalization in Trigonometric Polynomial Models
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 2319
EP - 2329
AU - Masashi SUGIYAMA
AU - Hidemitsu OGAWA
PY - 2001
DO -
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E84-A
IS - 9
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - September 2001
AB - In this paper, we consider the problem of active learning, and give a necessary and sufficient condition of sample points for the optimal generalization capability. By utilizing the properties of pseudo orthogonal bases, we clarify the mechanism of achieving the optimal generalization capability. We also show that the condition does not only provide the optimal generalization capability but also reduces the computational complexity and memory required to calculate learning result functions. Based on the optimality condition, we give design methods of optimal sample points for trigonometric polynomial models. Finally, the effectiveness of the proposed active learning method is demonstrated through computer simulations.
ER -