Comparison and Analysis for the Performance of Deep Learning-Based Time Series Prediction Algorithms According to Increasing Model Size 


Vol. 48,  No. 1, pp. 123-128, Jan.  2023
10.7840/kics.2023.48.1.123


PDF
  Abstract

This paper deals with empirical research on if scaling laws of language models are be able to be applied to state-of-the-art deep learning-based time series forecasting algorithms. Two deep learning-based time series forecasting algorithms have increased the size of the model by up to 24 times over those created in the original papers, and empirical analysis shows up to 3.6% better than the state-of-the-art performance of one of the algorithms. In addition, similar to the results of the scaling law paper on language models, the results of this paper confirm that model size and data size have a meaningful relationship with the performance of time series forecasting algorithms.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Related Articles
  Cite this article

[IEEE Style]

Y. Choi and D. Kim, "Comparison and Analysis for the Performance of Deep Learning-Based Time Series Prediction Algorithms According to Increasing Model Size," The Journal of Korean Institute of Communications and Information Sciences, vol. 48, no. 1, pp. 123-128, 2023. DOI: 10.7840/kics.2023.48.1.123.

[ACM Style]

Youngjoon Choi and Daekeun Kim. 2023. Comparison and Analysis for the Performance of Deep Learning-Based Time Series Prediction Algorithms According to Increasing Model Size. The Journal of Korean Institute of Communications and Information Sciences, 48, 1, (2023), 123-128. DOI: 10.7840/kics.2023.48.1.123.

[KICS Style]

Youngjoon Choi and Daekeun Kim, "Comparison and Analysis for the Performance of Deep Learning-Based Time Series Prediction Algorithms According to Increasing Model Size," The Journal of Korean Institute of Communications and Information Sciences, vol. 48, no. 1, pp. 123-128, 1. 2023. (https://doi.org/10.7840/kics.2023.48.1.123)
Vol. 48, No. 1 Index