Improving Deep Learning Models Considering the Time Lags between Explanatory and Response Variables
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
A regression model represents the relationship between explanatory and response variables. In real life,explanatory variables often affect a response variable with a certain time lag, rather than immediately. Forexample, the marriage rate affects the birth rate with a time lag of 1 to 2 years. Although deep learning modelshave been successfully used to model various relationships, most of them do not consider the time lags betweenexplanatory and response variables. Therefore, in this paper, we propose an extension of deep learning models,which automatically finds the time lags between explanatory and response variables. The proposed methodfinds out which of the past values of the explanatory variables minimize the error of the model, and uses thefound values to determine the time lag between each explanatory variable and response variables. After determiningthe time lags between explanatory and response variables, the proposed method trains the deep learningmodel again by reflecting these time lags. Through various experiments applying the proposed method to a fewdeep learning models, we confirm that the proposed method can find a more accurate model whose error isreduced by more than 60% compared to the original model.
Abstract
A regression model represents the relationship between explanatory and response variables. In real life,explanatory variables often affect a response variable with a certain time lag, rather than immediately. Forexample, the marriage rate affects the birth rate with a time lag of 1 to 2 years. Although deep learning modelshave been successfully used to model various relationships, most of them do not consider the time lags betweenexplanatory and response variables. Therefore, in this paper, we propose an extension of deep learning models,which automatically finds the time lags between explanatory and response variables. The proposed methodfinds out which of the past values of the explanatory variables minimize the error of the model, and uses thefound values to determine the time lag between each explanatory variable and response variables. After determiningthe time lags between explanatory and response variables, the proposed method trains the deep learningmodel again by reflecting these time lags. Through various experiments applying the proposed method to a fewdeep learning models, we confirm that the proposed method can find a more accurate model whose error isreduced by more than 60% compared to the original model.
참고문헌 (17)
- C. Kim, "Deep learning model for identifying the time lag between explanatory variables and response variable in regression analysis", Proceedings of Annual Conference of the Korea Information Processing Society, pp.868~871 (2021)
- J. Wu, "Prediction of birth rate in China under three-child policy based on neural network", Proceedings of 2022 7th International Conference on Intelligent Computing and Signal Processing (ICSP), pp.1652~1655 (2022)
- K. Kim, "Scenario analysis of fertility in Korea using the fertility rate prediction model", The Korean Journal of Applied Statistics, 28(4), pp.685~701 (2015)[DOI]
- T. Hayduk, "The effect of advertising on sales and brand equity in small sport businesses", Sport Marketing Quarterly, 30(3), pp.178~192 (2021)[DOI]
- Q. Sun, "A weighted-time-lag method to detect lag vegetation response to climate variation : a case study in Loess Plateau, China, 1982–2013", Remote Sensing, 13(5) (2021)[DOI]
- C. Szegedy, "Going deeper with convolutions", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.1~9 (2015)
- H. Gu, "Blind channel identification aided generalized automatic modulation recognition based on deep learning", IEEE Access, 7, pp.110722~110729 (2019)[DOI]
- M. Aamir, "An optimized architecture of image classification using convolutional neural network", International Journal of Image, Graphics and Signal Processing, 11(10), pp.30~39 (2019)[DOI]
- S. Miao, "A CNN regression approach for real-time 2D", 3D registration, 2016, pp.35(5), 1352~1363
- S. Hochreiter, "Long short-term memory", Neural Computation, 9(8), pp.1735~1780 (1997)[DOI]
- I. O. Tolstikhin, "MLP-mixer : an all-MLP architecture for vision", Advances in Neural Information Processing Systems, 34, pp.24261~24272 (2021)
- H. Han, "Residual learning based CNN for gesture recognition in robot interaction", Journal of Information Processing Systems, 17(2), pp.385~398 (2021)[DOI]
- J. Kim, "The method for generating recommended candidates through prediction of multi-criteria ratings using CNN-BiLSTM", Journal of Information Processing Systems, 17(4), pp.707~720 (2021)[DOI]
- "Binarized neural networks: training deep neural networks with weights and activations constrained to+ 1 or -1" (2016)[URL]
- J. Han, "Data Mining: Concepts and Techniques" (2001), Morgan Kaufmann
- S. Gold, "Softmax to softassign : neural network algorithms for combinatorial optimization", Journal of Artificial Neural Networks, 2(4), pp.381~399 (1996)
- "KOSIS (Korean Statistical Information Service)" (2024)[URL]