Abstract
In this paper, we discuss optimization of parameters for generation of sparse AutoEncoder based on the genetic algorithm, and show an efficient ensemble learning algorithm. From experiment result of some data sets, we generated better sparse AutoEncoder, and get ensemble effect from the diversity of data in the middle layer of AutoEncoder. Keywords: Sparse AutoEncoder, Ensemble Learning, Genetic Algorithm
Publisher
International Information Institute
Reference9 articles.
1. [1] Feng Xuanang, Genetic Algorithm-based Optimization of a Deep Neural Networks Ensemble and Its Applications, Department of Complex Systems Science, Graduate School of Informatics, Nagoya University, 2020-03-25.
2. [2] Diederik P. Kingma, Jimmy Lei Ba, ADAM: A Method for Stochastic Optimization, Conference paper at ICLR, 2015.
3. [3] Zhi-Hua Zhou, Ensemble Methods: Foundations and Algorithms, Chapman & Hall/CRC Machine Learning & Pattern Recognition Series, 2012.
4. [4] G. W. Cottrell and P. Munro, Principal component analysis of image via back-propagation, In Proceedings of SPIE 1001 Visual Communications and Image prossessing'88, pp. 1070-1076. 1988.
5. [5] L. K. Hansen, P. Salamon, Neural network ensembles, IEEE Transactions on Pattern Analysis and Machine Intelligence, 12(10): pp.993-1001, 1990.