The Journal of the Korea institute of electronic communication sciences (한국전자통신학회논문지)

Korea Institute of Electronic Communication Science (한국전자통신학회)
권호 표지
DOI QR코드

DOI QR Code

A Study on Machine Learning Algorithms based on Embedded Processors Using Genetic Algorithm

유전 알고리즘을 이용한 임베디드 프로세서 기반의 머신러닝 알고리즘에 관한 연구

  • 이소행 (순천대학교 SW중심대학사업단) ;
  • 석경휴 (순천대학교 SW중심대학사업단)
  • Received : 2024.02.03
  • Accepted : 2024.04.12
  • Published : 2024.04.30
Download PDF
⟨ Previous Next ⟩

Abstract

In general, the implementation of machine learning requires prior knowledge and experience with deep learning models, and substantial computational resources and time are necessary for data processing. As a result, machine learning encounters several limitations when deployed on embedded processors. To address these challenges, this paper introduces a novel approach where a genetic algorithm is applied to the convolution operation within the machine learning process, specifically for performing a selective convolution operation.In the selective convolution operation, the convolution is executed exclusively on pixels identified by a genetic algorithm. This method selects and computes pixels based on a ratio determined by the genetic algorithm, effectively reducing the computational workload by the specified ratio. The paper thoroughly explores the integration of genetic algorithms into machine learning computations, monitoring the fitness of each generation to ascertain if it reaches the target value. This approach is then compared with the computational requirements of existing methods.The learning process involves iteratively training generations to ensure that the fitness adequately converges.

일반적으로 머신러닝을 수행하기 위해서는 딥러닝 모델에 대한 사전 지식과 경험이 필요하고, 데이터를 연산하기 위해 고성능 하드웨어와 많은 시간이 필요하게 된다. 이러한 이유로 머신러닝은 임베디드 프로세서에서 실행하기에는 많은 제약이 있다.본 논문에서는 이러한 문제를 해결하기 위해 머신러닝의 과정 중 콘볼루션 연산(Convolution operation)에 유전 알고리즘을 적용하여 선택적 콘볼루션 연산(Selective convolution operation)과 학습 방법을 제안한다. 선택적 콘볼루션 연산에서는 유전 알고리즘에 의해 추출된 픽셀에 대해서만 콘볼루션을 수행하는 방식이다. 이 방식은 유전 알고리즘에서 지정한 비율만큼 픽셀을 선택하여 연산하는 방식으로 연산량을 지정된 비율만큼 줄일 수 있다. 본 논문에서는 유전 알고리즘을 적용한 머신러닝 연산의 심화학습을 진행하여 해당 세대의 적합도가 목표치에 도달하는지 확인하고 기존 방식의 연산량과 비교한다. 적합도가 충분히 수렴할 수 있도록 세대를 반복하여 학습하고, 적합도가 높은 모델을 유전 알고리즘의 교배와 돌연변이를 통해 다음 세대의 연산에 활용한다.

Keywords

Acknowledgement

본 연구는 2024년도 과학기술정보통신부 및 정보통신기획평가원의 SW중심대학사업 지원을 받아수행되었음. (2023-0-00028)

References

  1. Graziano Crasta, and V. De Cicco, "A chain rule formula in BV and applications to conservation laws", HomeSIAM Journal on Mathematical Analysis 3, 2010, vol 43, pp. 430-456. https://doi.org/10.1137/100804462
  2. John H. Holland, "Genetic Algorithms," Scientific American, 1992, vol. 267, no. 1, pp. 66-73. https://doi.org/10.1038/scientificamerican0792-66
  3. A. M. Saxe, J. L. McClelland, and S. Ganguli. "Exact solutions to the nonlinear dynamics of learning in deep linear neural networks." Proceedings of the International Conference on Learning Represenatations 2014, pp. 89-112.
  4. P. Sermanet, D. Eigen, X. Zhang, M. Mathieu, R. Fergus, and Y. LeCun. "Overfeat: Integrated recognition, localization and detection using convolutional networks." International Conference on Learning Representations, 2014, pp. 78-94.
  5. S. Lawrence, C. Lee Giles, A. C. Tsoi, and A. D. Back, "Face recognition : a convolutional neural network approach," IEEE Transactions on Neural Networks, 1997, vol. 8, no. 1, pp. 98-113. https://doi.org/10.1109/72.554195
  6. Y. Bengio and P. Frasconi, "Input-Output HMMs for sequence processing," IEEE Transactions on Neural Networks, 1996, vol. 7, no. 5, pp. 1231-1249. https://doi.org/10.1109/72.536317
  7. S.H. Kim, Y. G. Kim and W.J. Kim, "The Design of Method for Efficient Processing of Small Files in the Distributed System based on Hadoop Framework," J. of the Korea Institgrte of Electronic Communication Sciences, vol. 10. no. 10,2015, pp.1115-1121. https://doi.org/10.13067/JKIECS.2015.10.10.1115
  8. J. Deng, W. Dong, R. Socher, "ImageNet : A Large-scale hierarchical image database," IEEE, 2009, pp. 245-255.
  9. H. Lee, R. Grosse, R. Ranganath, and A.Y. Ng. "Convolutional deep belief networks for scalable unsupervised learning of hierarchical  representations," In Proceedings of the 26th Annual International Conference on Machine Learning of the ACM, 2009, pp. 609-616. 
  10. Krizhevsky, Alex, Ilya Sutskever, and Geoffrey E. Hinton. "Imagenet classification with deep convolutional neural networks." Communications of the ACM, 2017, pp. 84-90.
  11. Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun, "Deep residual learning for image recognition," IEEE, pp. 770-778, 2015
  12. Y. Bengio, P. Simard, and P. Frasconi., "Learning long-term dependentcies with gradient descent is difficult," IEEE Transactions on Neural Networks, 1994, vol. 5, no. 2, pp. 157-166. https://doi.org/10.1109/72.279181
  13. P. Sermanet, D. Eigen, X. Zhang, M. Mathieu, R. Fergus, and Y. LeCun. "Overfeat: Integrated recognition, localization and detection using convolutional networks." International Conference on Learning Representations, 2014, pp. 78-94