Simulating Time-Series Data for Improved Deep Neural Network Performance
| Autor: | David Booth, Dominic Thewlis, Brian W.-H. Ng, Jordan Yeomans, William S. P. Robertson, Simon Thwaites |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
General Computer Science
Computer science 02 engineering and technology Overfitting Machine learning computer.software_genre 03 medical and health sciences 0302 clinical medicine 0202 electrical engineering electronic engineering information engineering General Materials Science Time series data simulation Series (mathematics) Artificial neural network business.industry Deep learning 020208 electrical & electronic engineering SIGNAL (programming language) General Engineering time-series data augmentation time-series classification ComputingMethodologies_PATTERNRECOGNITION deep neural networks Labeled data lcsh:Electrical engineering. Electronics. Nuclear engineering Artificial intelligence business lcsh:TK1-9971 computer 030217 neurology & neurosurgery data augmentation |
| Zdroj: | IEEE Access, Vol 7, Pp 131248-131255 (2019) |
| ISSN: | 2169-3536 |
| DOI: | 10.1109/access.2019.2940701 |
| Popis: | Deep learning algorithms have shown remarkable performance in classification tasks, however, they typically perform poorly with small training datasets due to overfitting. Overfitting occurs for all data types, although for the purposes of this study we are interested in time-based signals. This study introduces a novel technique to simulate time series signals from a dataset of categorically labeled data which can be used to train a deep neural network. The objective is to improve the predictive accuracy of a deep neural network on a separate validation dataset. To demonstrate the simulation methodology and improvements to the model's performance, a small dataset of ground reaction forces was used with the goal of identifying a person based on the raw signal. Our results show that the simulation method presented improves validation accuracy and reduces model training time for each of the three signal types. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|