실천공학교육논문지 (Journal of Practical Engineering Education)

  • 제14권3호
  • /
  • Pages.491-498
  • /
  • 2022
  • /
  • 2288-405X(pISSN)
  • /
  • 2288-4068(eISSN)
한국실천공학교육학회 (Korean Institute for Pratical Engineering Education)
권호 표지
DOI QR코드

DOI QR Code

순환신경망 기초 실습 사례 개발

Development of Basic Practice Cases for Recurrent Neural Networks

  • 허경 (경인교육대학교 컴퓨터교육과)
  • Kyeong Hur (Department of Computer Education, Gyeong-In National University of Education)
  • 투고 : 2022.11.27
  • 심사 : 2022.12.07
  • 발행 : 2022.12.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문에서는 비전공자들을 위한 교양과정으로, 기초 순환신경망 과목 커리큘럼을 설계하는데 필수적으로 요구되는 순환신경망 SW 실습 사례를 개발하였다. 개발된 SW 실습 사례는 순환신경망의 동작원리를 이해시키는 데 초점을 두고, 시각화된 전체 동작 과정을 확인할 수 있도록 스프레드시트를 사용하였다. 개발된 순환신경망 실습 사례는 지도학습 방식의 텍스트완성 훈련데이터 생성, 입력층, 은닉층, 상태층(컨텍스트 노드) 그리고 출력층을 차례대로 구현하고, 텍스트 데이터에 대해 순환신경망의 성능을 테스트하는 것으로 구성되었다. 본 논문에서 개발한 순환신경망 실습사례는 다양한 문자 수를 갖는 단어를 자동 완성한다. 제안한 순환신경망 실습사례를 활용하여, 한글 또는 영어 단어를 구성하는 최대 문자 수를 다양하게 확장하여 자동 완성하는 인공지능 SW 실습 사례를 만들 수 있다. 따라서, 본 순환신경망 기초 실습 사례의 활용도가 높다고 할 수 있다.

In this paper, as a liberal arts course for non-major students, a case study of recurrent neural network SW practice, which is essential for designing a basic recurrent neural network subject curriculum, was developed. The developed SW practice case focused on understanding the operation principle of the recurrent neural network, and used a spreadsheet to check the entire visualized operation process. The developed recurrent neural network practice case consisted of creating supervised text completion training data, implementing the input layer, hidden layer, state layer (context node), and output layer in sequence, and testing the performance of the recurrent neural network on text data. The recurrent neural network practice case developed in this paper automatically completes words with various numbers of characters. Using the proposed recurrent neural network practice case, it is possible to create an artificial intelligence SW practice case that automatically completes by expanding the maximum number of characters constituting Korean or English words in various ways. Therefore, it can be said that the utilization of this case of basic practice of recurrent neural network is high.

키워드

참고문헌

  1. Y. Bengio, I. Goodfellow, and A. Courville, "Deep learning," MIT Press, 2017.
  2. H. Kurt, M. Stinchcombe, and H. White1, "Multilayer feedforward networks are universal approximators," Neural Networks, vol. 2, issue 5, pp. 359-366, 1989. https://doi.org/10.1016/0893-6080(89)90020-8
  3. M. Roodschild, J. Gotay Sardinas, and A. Will, "A new approach for the vanishing gradient problem on sigmoid activation," Progress in Artificial Intelligence, vol. 9, no. 4, pp. 351-360, 2020. https://doi.org/10.1007/s13748-020-00218-y
  4. V. Nair and G. Hinton, "Rectified linear units improve restricted boltzmann machines," International Conference on Machine Learning, pp. 807-814, 2010.
  5. Y. Qin, X. Wang, and J. Zou, "The optimized deep belief networks with improved logistic Sigmoid units and their application in fault diagnosis for planetary gearboxes of wind turbines," IEEE Transactions on Industrial Electronics, vol. 66, no. 5, pp. 3814-3824, July 2018.
  6. X. Wang, Y. Qin, Y. Wang, S. Xiang, and H. Chen, "ReLTanh: An activation function with vanishing gradient resistance for SAE-based DNNs and its application to rotating machinery fault diagnosis," Neurocomputing, vol. 363, pp. 88-98, 2019. https://doi.org/10.1016/j.neucom.2019.07.017
  7. S. Kong and M. Takatsuka, "Hexpo: A vanishing-proof activation function," International Joint Conference on Neural Networks, pp. 2562-2567, 2017.
  8. Wakui Yoshiyuki and Wakui Sadami, "Introduction to recurrent neural networks and reinforcement learning in excel," Seoul : Seongandang, 2020.
  9. K. Hur, "Supervised learning artificial neural network parameter optimization and activation function basic training method using spreadsheets," Journal of Practical Engineering Education, vol. 13, no. 2, pp. 233-242, 2021.