Communication Papers of the 20th Conference on Computer Science and Intelligence Systems (FedCSIS)

DOI: http://dx.doi.org/10.15439/2025F1427

Citation: Communication Papers of the 20th Conference on Computer Science and Intelligence Systems (FedCSIS), M. Bolanowski, M. Ganzha, L. Maciaszek, M. Paprzycki, D. Ślęzak (eds). ACSIS, Vol. 45, pages 41–47 (2025)

Full text

Abstract. In this paper, we present a framework that integrates Particle Swarm Optimization (PSO), machine learning, K-Fold cross-validation, and surrogate modeling to identify the optimal weight vector for feature scaling in neural network training. The n-dimensional weight space is first partitioned into non-overlapping subdomains, each treated as a PSO particle. Particle movement is driven by a characteristic vector---derived from the top-performing candidates within each subdomain---and by information exchanged with neighboring subdomains. To reduce evaluation costs, a surrogate model trained on a uniformly sampled subset of candidates is used to pre-filter particles before full K-Fold validation. The top performers then undergo comprehensive validation, updating the characteristic vectors for subsequent iterations. Our domain-as particle PSO framework enables efficient weight discovery with significant reductions in computational overhead while maintaining robust performance. The effectiveness of this approach has been demonstrated on real-world datasets.

References

1. Katharina Benz, Claus Rech, Paolo Mercorelli, Sustainable management of marine fish stocks by means of sliding mode control, in Proceedings of the 2019 Federated Conference on Computer Science and Information Systems, Volume 18 of FedCSIS 2019. year, IEEE, September 2019, ISSN 2300-5963, https://dx.doi.org/10.15439/2019f221.
2. Katharina Benz, Claus Rech, Paolo Mercorelli, Oleg Sergiyenko, Two cascaded and extended Kalman filters combined with sliding mode control for sustainable management of marine fish stocks, Journal of Automation, Mobile Robotics and Intelligent Systems, p. 28–35, July 2019, ISSN 2080-2145, http://dx.doi.org/10.14313/JAMRIS/3-2020/30.
3. J. Kennedy, R. Eberhart, Particle swarm optimization, in Proceedings of ICNN’95 - International Conference on Neural Networks, year, Volume 4, 1995, https://dx.doi.org/10.1109/ICNN.1995.488968.
4. Piotr Andrzej Kowalski, Maciej Kusy, Determining the significance of features with the use of sobol’ method in probabilistic neural network classification tasks, year, in Proceedings of the 2017 Federated Conference on Computer Science and Information Systems (FedCSIS), 2017, https://dx.doi.org/10.15439/2017F225.
5. Dzhakhongir Normatov, Paolo Mercorelli, Parameters estimation of a Lotka-Volterra model in an application for market graphics processing units, year, in 2022 17th Conference on Computer Science and Intelligence Systems (FedCSIS), 2022, https://dx.doi.org/10.15439/2022F61.
6. Tareq M. Shami, Ayman A. El-Saleh, Mohammed Alswaitti, Qasem Al-Tashi, Mhd Amen Summakieh, Seyedali Mirjalili, Particle swarm optimization: A comprehensive survey, IEEE Access, 10:10031–10061, 2022, https://dx.doi.org/10.1109/ACCESS.2022.3142859.
7. Bing Xue, Mengjie Zhang, Will N. Browne, Xin Yao, A survey on evolutionary computation approaches to feature selection, IEEE Transactions on Evolutionary Computation, 20(4):606–626, 2016, https://dx.doi.org/10.1109/TEVC.2015.2504420.
8. Yudong Zhang, Lenan Wu, Shuihua Wang, Magnetic resonance brain image classification by an improved artificial Bee Colony Algorithm, Progress In Electromagnetics Research, 116:65–79, 2011, ISSN 1559- 8985, https://dx.doi.org/10.2528/pier11031709.