Hyper Parameter Tuning Method based on Sampling for Optimal LSTM Model

KIM HYEMEE; RyeJi Jeong; BAE HYE RIM

doi:https://doi.org/10.9708/jksci.2019.24.01.137

Journal of The Korea Society of Computer and Information

Abbr : JKSCI
2019, 24(1), pp.137-143
DOI :https://doi.org/10.9708/jksci.2019.24.01.137
Publisher : The Korean Society Of Computer And Information
Research Area : Computer Science

KIM HYEMEE ¹ , RyeJi Jeong ² , BAE HYE RIM ³

¹ 부산대학교
² 부산대학교
³ 부산대학교

ABSTRACT

As the performance of computers increases, the use of deep learning, which has faced technical limitations in the past, is becoming more diverse. In many fields, deep learning has contributed to the creation of added value and used on the bases of more data as the application become more divers. The process for obtaining a better performance model will require a longer time than before, and therefore it will be necessary to find an optimal model that shows the best performance more quickly. In the artificial neural network modeling a tuning process that changes various elements of the neural network model is used to improve the model performance. Except Gride Search and Manual Search, which are widely used as tuning methods, most methodologies have been developed focusing on heuristic algorithms. The heuristic algorithm can get the results in a short time, but the results are likely to be the local optimal solution. Obtaining a global optimal solution eliminates the possibility of a local optimal solution. Although the Brute Force Method is commonly used to find the global optimal solution, it is not applicable because of an infinite number of hyper parameter combinations. In this paper, we use a statistical technique to reduce the number of possible cases, so that we can find the global optimal solution.
KEYWORDS

LSTM , Optimization , Hyper parameter tuning , Sampling
REFERENCES (16)

[journal] L. Deng / 2014 / Deep learning : methods and applications / Foundations and Trends in Signal Processing 7 (3-4) : 197~387

[book] C. Paar / 2009 / Understanding cryptography : a textbook for students and practitioners / Springer Science & Business Media : 7~

[journal] R. E. Korf / 1985 / Depth-first iterative-deepening: An optimal admissible tree search / Artificial intelligence 27 (1) : 97~109

[journal] T. Pukkala / 1993 / A heuristic optimization method for forest planning and decision making / Scandinavian Journal of Forest Research 8 (1-4) : 560~570

[book] A. Graves / 2012 / Supervised sequence labelling with recurrent neural networks / Springer : 35~42

[confproc] R. Socher / 2013 / Recursive deep models for semantic compositionality over a sentiment treebank / Empirical Methods in Natural Language Processing : 1631~1642

[confproc] T. Mikolov / 2010 / Recurrent neural network based language model / the International Speech Communication Association

[journal] S. Hochreiter / 1997 / Long short-term memory / Neural computation 9 (8) : 1735~1780

[book] R. V. Hogg / 2015 / Probability and statistical inference / Pearson Educational International : 204~205

[journal] J. Bergstra / 2012 / Random search for hyper-parameter optimization / Journal of Machine Learning Research 13 : 281~305

[confproc] H. K. Lam / 2001 / Tuning of the structure and parameters of neural network using an improved genetic algorithm / Industrial Electronics Society 2001(IECON’01) : 25~30

[confproc] J. Snoek / 2012 / Practical bayesian optimization of machine learning algorithms / Advances in neural information processing systems : 2951~2959

[journal] Y. Xia / 2017 / A boosted decision tree approach using Bayesian hyper-parameter optimization for credit scoring / Expert Systems with Applications 78 : 225~241

[confproc] B. Qolomany / 2017 / Parameters optimization of deep learning models using particle swarm optimization / Wireless Communications and Mobile Computing Conference(IWCMC) : 1285~1290

[journal] A. Tharwat / 2017 / A ba-based algorithm for parameter optimization of support vector machine / Pattern Recognition Letters 93 : 13~22

[journal] C. Zhang / 2017 / A compound structure of ELM based on feature selection and parameter optimization using hybrid backtracking search algorithm for wind speed forecasting / Energy Conversion and Management 143 : 360~376
KCI Citation (0)

KJC Korea
Journal Central

Statistics

Tools