Artificial Neural Network Approach for Quantum Defect Prediction in Alkali Atoms

Abstract

In this study, an Artificial Neural Network (ANN) model was suggested and trained to predict the quantum defect values of alkali atoms. The dataset was divided into training and testing subsets with a % 60 – % 40, respectively. To prevent overfitting, the number of training epochs was limited to 250, and the learning rate was set to 0.25. The training process employed the Gradient Descent optimization algorithm for updating the network weights. Two different activation functions, ReLU and Swish, were utilized to evaluate their impact on prediction accuracy. The predicted quantum defect values obtained from the ANN were compared with corresponding experimental results to assess the model’s performance.

Keywords

• artificial neural network
• gradiend descent
• alkali atoms

Alkali Atomlarda Kuantum Kusur Tahmini İçin Yapay Sinir Ağı Yaklaşımı

Abstract

Bu çalışmada, alkali atomların kuantum defekt değerlerini tahmin etmek amacıyla bir Yapay Sinir Ağı (YSA) modeli önerilmiş ve eğitilmiştir. Veri seti, sırasıyla %60 eğitim ve %40 test alt kümelerine ayrılmıştır. Aşırı öğrenmeyi önlemek amacıyla eğitim epoch sayısı 250 ile sınırlandırılmış ve öğrenme oranı 0.25 olarak belirlenmiştir. Ağ ağırlıklarının güncellenmesinde Gradient Descent optimizasyon algoritması kullanılmıştır. Tahmin doğruluğu üzerindeki etkilerini değerlendirmek amacıyla ReLU ve Swish olmak üzere iki farklı aktivasyon fonksiyonu kullanılmıştır. YSA ile elde edilen kuantum defekt tahminleri, karşılık gelen deneysel sonuçlarla karşılaştırılarak modelin performansı değerlendirilmiştir.

Keywords

• Yapay sinir ağları
• gradyan inişi
• alkali atomlar

References

1. Schmidgall, S., et al. (2024). Brain-inspired learning in artificial neural networks: A review. APL Machine Learning, 2(2), 021501. https://doi.org/10.1063/5.0186054
2. Moustris, K. P., et al. (2013). Development and application of artificial neural network mo-deling in forecasting PM10 levels in a Mediterranean city. Water Air & Soil Pollution, 224(8), 1634. https://doi.org/10.1007/s11270-013-1634-x.
3. Grossi, E., and Buscema, M. (2007). Introduction to artificial neural networks. European Journal of Gastroenterology & Hepatology, 19(12), 1046-1054. https://doi.org/10.1097/MEG.0b013e3282f198a0
4. Hopfield, J. J. (1988). Artificial neural networks. IEEE Circuits & Devices Magazine, 4(5), 3-10. https://doi.org/10.1109/101.8118.
5. Abiodun, O. I., et al. (2018). State of the art in artificial neural network applications: A survey. Heliyon, 4(11), e00938. https://doi.org/10.1016/j.heliyon.2018.e00938.
6. Duch, W., and Diercksen, G. H. (1994). Neural networks as tools to solve problems in physics and chemistry. Computational Physics Communications, 82(2-3), 91-103. https://doi.org/10.1016/0010-4655(94)90158-9.
7. Denby, B. (1993). The use of neural networks in high-energy physics. Neural Computation, 5(4), 505-549. https://doi.org/10.1162/neco.1993.5.4.505.
8. Forte, S. (2024). On becoming the new editor in chief of J Phys G: Nucl Part Phys. Journal of Physics G: Nuclear and Particle Physics, 51(5), 050201. https://doi.org/10.1088/1361-6471/ad33e3.

9. Pang, L. G. (2024). Studying high-energy nuclear physics with machine learning. International Journal of Modern Physics E, 33(6), 2430009. https://doi.org/10.1142/S0218301324300091.
10. Kara, S. O., and Akkoyun, S. (2024). A search for leptonic photon, Zl, at all three CLIC energy stages by using artificial neural networks (ANN). Acta Physica Polonica B, 55(6), 1. https://doi.org/10.5506/APhysPolB.55.6-A4.
11. Raza, S., et al. (2024). Prediction of spectral characteristics of lithium-like ions by artificial neural network. Indian Journal of Physics. https://doi.org/10.1007/s12648-024-03346-6.
12. Rau, A. R. P., and Inokuti, M. (1997). The quantum defect: Early history and recent deve-lopments. American Journal of Physics, 65(3), 221-225. https://doi.org/10.1119/1.18532.
13. Bergmann, L., and Schaefer, C. (1997). Constituents of matter: Atoms, molecules, nuclei, and particles. Walter de Gruyter.
14. Qamar, R., and Zardari, B. A. (2023). Artificial neural networks: An overview. Mesopota-mian Journal of Computer Science, 130-139. https://doi.org/10.58496/MJCSC/2023/015.
15. Sola, J., and Sevilla, J. (1997). Importance of input data normalization for the application of neural networks to complex industrial problems. IEEE Transactions on Nuclear Science, 44(3), 1464-1468. https://doi.org/10.1109/23.589532.
16. Huang, L., et al. (2023). Normalization techniques in training DNNs: Methodology, analysis, and application. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(8), 10173-10196. https://doi.org/10.1109/TPAMI.2023.3250241.
17. Department of CSE, VIIT(A), AP, India University, India, et al. (2020). A study on backpropagation in artificial neural networks. Asia-Pacific Journal of Neural Networks and Its Applications, 4(1), 21-28. https://doi.org/10.21742/AJNNIA.2020.4.1.03.
18. Wilson, D. R., and Martinez, T. R. (2001). The need for small learning rates on large prob-lems. Proceedings of the IJCNN'01 International Joint Conference on Neural Networks, 1, 115-119. https://doi.org/10.1109/IJCNN.2001.939002.
19. Dongare, A. D. D., Kharde, R. R., and Kachare, D. K. (2012). Introduction to artificial neural network.
20. Mitra, M., et al. (2019). Optimizing neural networks: A comparative analysis of activation functions in deep learning. International Journal of Scientific Research, 8(3), 1940-1949. https://doi.org/10.21275/SR231205140623.
21. Kingma, D., and Ba, J. (2015). Adam: A method for stochastic optimization. Proceedings of the 3rd International Conference on Learning Representations (ICLR 2015).
22. Kumar, S. (2017). On weight initialization in deep neural networks. arXiv:1704.08863.