TY - GEN
T1 - Improvement algorithm for approximate incremental learning
AU - Oyama, Tadahiro
AU - Choge, H. Kipsang
AU - Karungaru, Stephen
AU - Tsuge, Satoru
AU - Mitsukura, Yasue
AU - Fukumi, Minoru
PY - 2009
Y1 - 2009
N2 - This paper presents an improved algorithm of Incremental Simple-PCA. The Incremental Simple-PCA is a fast incremental learning algorithm based on Simple-PCA. This algorithm need not hold all training samples because it enables update of an eigenvector according to incremental samples. Moreover, this algorithm has an advantage that it can calculate the eigenvector at high-speed because matrix calculation is not needed. However, it had a problem in convergence performance of the eigenvector. Thus, in this paper, we try the improvement of this algorithm from the aspect of convergence performance. We performed computer simulations using UCI datasets to verify the effectiveness of the proposed algorithm. As a result, its availability was confirmed from the standpoint of recognition accuracy and convergence performance of the eigenvector compared with the Incremental Simple-PCA.
AB - This paper presents an improved algorithm of Incremental Simple-PCA. The Incremental Simple-PCA is a fast incremental learning algorithm based on Simple-PCA. This algorithm need not hold all training samples because it enables update of an eigenvector according to incremental samples. Moreover, this algorithm has an advantage that it can calculate the eigenvector at high-speed because matrix calculation is not needed. However, it had a problem in convergence performance of the eigenvector. Thus, in this paper, we try the improvement of this algorithm from the aspect of convergence performance. We performed computer simulations using UCI datasets to verify the effectiveness of the proposed algorithm. As a result, its availability was confirmed from the standpoint of recognition accuracy and convergence performance of the eigenvector compared with the Incremental Simple-PCA.
KW - Cincremental learning Cdimensional reduction Cpattern recognition
KW - PCA
KW - Simple-PCA
UR - http://www.scopus.com/inward/record.url?scp=76649137602&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=76649137602&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-10677-4_59
DO - 10.1007/978-3-642-10677-4_59
M3 - Conference contribution
AN - SCOPUS:76649137602
SN - 3642106765
SN - 9783642106767
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 520
EP - 529
BT - Neural Information Processing - 16th International Conference, ICONIP 2009, Proceedings
T2 - 16th International Conference on Neural Information Processing, ICONIP 2009
Y2 - 1 December 2009 through 5 December 2009
ER -