Hybrid Multidimensional Wavelet-Neuro-System and its Learning Using Cross Entropy Cost Function in Pattern Recognition
| Autor: | Yuriy Borzov, Semen Oskerko, Viktor Voloshyn, Olena Vynokurova, Dmytro Peleshko |
|---|---|
| Přispěvatelé: | IT Step University, Lviv State University of Life Safety |
| Rok vydání: | 2018 |
| Předmět: |
Computer science
business.industry Approximation algorithm patterns recognition Pattern recognition learning algorithm Statistical classification Cross entropy Wavelet Task analysis Entropy (information theory) hybrid wavelet-neurosystem Artificial intelligence business wavelet transform cross entropy cost function |
| Zdroj: | 2018 IEEE Second International Conference on Data Stream Mining & Processing (DSMP). |
| DOI: | 10.1109/dsmp.2018.8478608 |
| Popis: | In this paper, the hybrid multidimensional wavelet-neuro-system for pattern recognition tasks is proposed. Also learning algorithm for all its parameters (synaptic weights, the centers, and widths of wavelet activation functions) based on cross entropy cost function was proposed. The proposed system is characterized by high learning speed and high approximation properties in comparison with well-known approaches. The efficiency of the proposed approach has been justified based on different benchmarks and real data sets. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|