Journals / Gazi Mühendislik Bilimleri Dergisi / 2023 / Cilt: 9 Sayı: 2

Storage of Developed Distributed Machine Learning Models on Blockchain

Geliştirilen Dağıtılmış Makine Öğrenimi Modellerinin Blok Zincirde Depolanması

Pages
248–263
Publication Date
2023-08-31
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

For this reason, a robust blockchain-based approach is proposed against malicious attacks during model deployment.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:In this study, a data set containing air pollution information collected from different locations was used.
📊
Key Findings & Data:This dataset contains air pollution measurement information and results information for 25 different districts of Seoul city.
💡
Conclusion & Impact:For this reason, a robust blockchain-based approach is proposed against malicious attacks during model deployment.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Dağıtık öğrenme · makine öğrenmesi · blok zinciri

Abstract

In this study, a data set containing air pollution information collected from different locations was used. This dataset contains air pollution measurement information and results information for 25 different districts of Seoul city. The values obtained from similar regions were filtered from the entire data set and a data set specific to these regions was created. As the next step, the model was trained with the values obtained in its own regions and the model was recorded. When this model is tested with the data obtained in its own fields, it shows very successful results. However, when tested with data obtained from different regions and dissimilar to these regions, the model classification success remained very low. The aim of the study is to show that more successful results can be obtained with the distributed learning method, which can be a solution to this problem situation, compared to classical machine learning. In addition, security problems during model aggregation and aggregation should not be ignored in the Distributed Learning section. For this reason, a robust blockchain-based approach is proposed against malicious attacks during model deployment.

Özet

Bu çalışmada farklı lokasyonlardan toplanan hava kirliliği bilgilerini içeren bir veri seti kullanılmıştır. Bu veri seti Seul şehrinin 25 farklı ilçesine ait hava kirliliği ölçüm bilgileri ve sonuç bilgilerini içermektedir. Benzer bölgelerden elde edilen değerler tüm veri seti içerisinden filtrelenerek bu bölgelere has veri seti oluşturulmuştur. Sonraki işlem basamağı olarak, kendi bölgelerinde elde edilen değerler ile eğitilip model kaydedilmiştir. Bu model kendi alanlarında elde edilen veriler ile test edildiğinde oldukça başarılı sonuçlar ortaya koymaktadır. Fakat farklı bölgelerden elde edilen ve bu bölgeler ile benzeşmeyen veriler ile test edildiğinde model sınıflama başarısı oldukça düşük kalmıştır. Çalışmanın amacı bu problem durumuna çözüm olabilecek olan dağıtık öğrenme yöntemi ile klasik makine öğrenmesine kıyasla daha başarılı sonuçlar elde edilebileceğini göstermektir. Bunun yanında Dağıtık Öğrenme kısmında model birleştirme ve toplama esnasındaki güvenlik sorunlarının da göz ardı edilmemelidir. Bu sebeple model dağıtımı sırasında kötü niyetli saldırılara karşı blockchain tabanlı sağlam bir yaklaşım önerilmiştir.

🔖

Cite This Article

GÜRFİDAN, R., ERSOY, M. (2023). Storage of Developed Distributed Machine Learning Models on Blockchain. Gazi Mühendislik Bilimleri Dergisi, Cilt: 9 Sayı: 2, 248–263. https://ajindex.com/articles/960503

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (1)

  1. O. Güler and S. Savaş, “All aspects of metaverse studies, technologies and future,” Gazi Journal of Engineering Sciences, vol. 8, no. 2, pp. 292–319, Sep. 2022. doi:10.30855/gmbd.0705011 [2] E. A. Çubukçu, V. Demir and M. F. Sevimli, “Estimating streamflow data with machine learning techniques,” Gazi Journal of Engineering Sciences, vol. 8, no. 2, pp. 257–272, Sep. 2022. doi:10.30855/gmbd.0705009 [3] K. Yang, T. Jiang, Y. Shi, and Z. Ding, “Federated learning via over-the-air computation,” IEEE Trans Wirel Commun, vol. 19, no. 3, pp. 2022–2035, Mar. 2020. doi:10.1109/TWC.2019.2961673 [4] T. Li, A. K. Sahu, A. Talwalkar, and V. Smith, “Federated learning: challenges, methods, and future directions,” IEEE Signal Process Mag, vol. 37, no. 3, pp. 50–60, May 2020. doi:10.1109/MSP.2020.2975749 [5] Cao, X., Başar, T., Diggavi, S., Eldar, Y. C., Letaief, K. B., Poor, H. V., & Zhang, J., “Communication-efficient distributed learning: an overview,” IEEE Journal on Selected Areas in Communications, vol. 41, no. 4, pp. 851–873, Apr. 2023. doi:10.1109/JSAC.2023.3242710 [6] Li, Q., Wen, Z., Wu, Z., Hu, S., Wang, N., Li, Y., Liu, X., He, B., “A survey on federated learning systems: vision, hype and reality for data privacy and protection,” IEEE Trans Knowl Data Eng, Apr. 2021. doi:10.1109/TKDE.2021.3124599 [7] C. Zhang, Y. Xie, H. Bai, B. Yu, W. Li, and Y. Gao, “A survey on federated learning,” Knowl Based Syst, vol. 216, p. 106775, Mar. 2021. doi:10.1016/J.KNOSYS.2021.106775 [8] J. P. Albrecht, “How the GDPR Will Change the World,” European Data Protection Law Review (EDPL), vol. 2, 2016, Accessed: Jun. 20, 2023. [Online]. Available: https://heinonline.org/HOL/Page?handle=hein.journals/edpl2&id=313&div=&collection= [9] X. Wang, W. Qin, F. Jiao, L. Dong, J. Guo, J. Zhang and C. Yang, “Review of tungsten resource reserves, tungsten concentrate production and tungsten beneficiation technology in China,” Transactions of Nonferrous Metals Society of China, vol. 32, no. 7, pp. 2318–2338, Jul. 2022. doi:10.1016/S1003-6326(22)65950-8 [10] S. Raschka, “MLxtend: Providing machine learning and data science utilities and extensions to Python’s scientific computing stack,” J Open Source Softw, vol. 3, no. 24, p. 638, Apr. 2018. doi:10.21105/joss.00638 [11] A. Gaier and D. Ha, “Weight agnostic neural networks,” Adv Neural Inf Process Syst, vol. 32, 2019, Accessed: Jun. 20, 2023. [Online]. Available: https://weightagnostic.github.io/ [12] E. Tijan, S. Aksentijević, K. Ivanić and