Observational Learning with Modular Networks
2000
Conference Paper
ei
Observational learning algorithm is an ensemble algorithm where each network is initially trained with a bootstrapped data set and virtual data are generated from the ensemble for training. Here we propose a modular OLA approach where the original training set is partitioned into clusters and then each network is instead trained with one of the clusters. Networks are combined with different weighting factors now that are inversely proportional to the distance from the input vector to the cluster centers. Comparison with bagging and boosting shows that the proposed approach reduces generalization error with a smaller number of networks employed.
Author(s): | Shin, H. and Lee, H. and Cho, S. |
Journal: | Lecture Notes in Computer Science (LNCS 1983) |
Volume: | LNCS 1983 |
Pages: | 126-132 |
Year: | 2000 |
Month: | July |
Day: | 0 |
Publisher: | Springer-Verlag |
Department(s): | Empirical Inference |
Bibtex Type: | Conference Paper (inproceedings) |
Event Name: | International Conference on Intelligent Data Engineering and Automated Learning (IDEAL) |
Event Place: | HK, China |
Address: | Heidelberg |
Digital: | 0 |
Institution: | Seoul National University, Seoul, Korea |
Organization: | Max-Planck-Gesellschaft |
School: | Biologische Kybernetik |
Links: |
PDF
|
BibTex @inproceedings{2691, title = {Observational Learning with Modular Networks}, author = {Shin, H. and Lee, H. and Cho, S.}, journal = {Lecture Notes in Computer Science (LNCS 1983)}, volume = {LNCS 1983}, pages = {126-132}, publisher = {Springer-Verlag}, organization = {Max-Planck-Gesellschaft}, institution = {Seoul National University, Seoul, Korea}, school = {Biologische Kybernetik}, address = {Heidelberg}, month = jul, year = {2000}, doi = {}, month_numeric = {7} } |