Bayesian neural networks and Gaussian processes in identification of concrete properties

• Autorzy: Słoński M.
• Języki publikacji: EN

Abstrakty

EN

This paper gives a concise overview of concrete properties prediction using advanced nonlinear regression approach and Bayesian inference. Feed-forward layered neural network (FLNN) with Markov chain Monte Carlo stochastic sampling and Gaussian process (GP) with maximum likelihood hyperparameters estimation are introduced and compared. An empirical assessment of these two models using two benchmark problems are presented. Results on these benchmark datasets show that Bayesian neural networks and Gaussian processes have rather similar prediction accuracy and are superior in comparison to linear regression model.

Słowa kluczowe

EN

nonlinear regression; Bayesian method; concrete; Gaussian process

PL

regresja nieliniowa; metoda bayesowska; beton

• Wydawca: Instytut Podstawowych Problemów Techniki PAN
• Czasopismo: Computer Assisted Mechanics and Engineering Sciences
• Rocznik: 2011
• Tom: Vol. 18, no. 4
• Strony: 291--302
• Opis fizyczny: Bibliogr. 18 poz., rys., tab., wykr.

Twórcy

autor

Słoński M.

• Cracow University of Technology Institute for Computational Civil Engineering Warszawska 24, 31-155 Kraków, Poland,

Bibliografia

• [1] C.M. Bishop. Pattern Recognition and Machine Learning. Springer, New York, 2006.
• [2] K. Furtak. Strength of the concrete under multiple repeated loads [in Polish]. Arch. of Civil Eng., 30, 1984.
• [3] S. Haykin. Neural Networks, a comprehensive foundation. Prentice Hall, 1999.
• [4] M. Jakubek and Z. Waszczyszyn. Neural analysis of concrete fatigue durability by the neuro-fuzzy FWNN. In L. Rutkowski, J. Siekmann, R. Tadeusiewicz, and Lotfi A. Zadeh [Eds.], Artificial Intelligence and Soft Computing – ICAISC 2004, Lecture Notes in Artificial Intelligence. Springer Berlin/Heidelberg, 2004.
• [5] J. Kaliszuk, A. Urbańska, Z. Waszczyszyn, and K. Furtak. Neural analysis of concrete fatigue durability on the basis of experimental evidence. Arch. of Civil Eng., 38, 2001.
• [6] J. Kasperkiewicz, J. Racz, and A. Dubrawski. HPC strength prediction using artificial neural network. Journal of Computing in Civil Engineering, 9(4): 1–6, 1995.
• [7] J. Lampinen and A. Vehtari. Bayesian approach for neural networks – review and case studies. Neural Networks, 14(3): 7–24, April 2001. (Invited article).
• [8] I.T. Nabney. Netlab: Algorithms for Pattern Recognition. Springer, London, 2002.
• [9] R.M. Neal. Bayesian training of backpropagation networks by the hybrid Monte Carlo method. Technical Report CRG-TR-92-1, 1992.
• [10] R.M. Neal. Bayesian Learning for Neural Networks. Lecture Notes in Statistics 118. Springer, 1996.
• [11] J.W. Oh, I.W. Lee, J.T. Kim, and G.W. Lee. Application of neural networks for proportioning of concrete mixes. ACI Materials Journal, 96: 61–67, 1999.
• [12] C.E. Rasmussen and C.K.I. Williams. Gaussian Processes for Machine Learning. The MIT Press, Cambridge, Massachusetts, 2006.
• [13] M. Słoński. Bayesian regression approaches on example of concrete fatigue failure prediction. Computer Assisted Mech. Eng. Sci., 13(4): 655–668, 2006.
• [14] M. Słoński. HPC strength prediction using Bayesian neural networks. Computer Assisted Mech. Eng. Sci., 14(1), 2007.
• [15] M. Słoński. A comparison of model selection methods for compressive strength prediction of high-performance concrete using neural networks. Computers & Structures, 88(21–22): 1248–1253, 2010.
• [16] A. Vehtari. MCMCstuff toolbox for MATLAB. User Manual, 2006.
• [17] Z. Waszczyszyn and M. Słoński. Some problems of artificial neural networks design. In Z. Waszczyszyn [Ed.], Advances of Soft Computing in Engineering, volume 512 of CISM Lectures and Notes, pages 237–316. Springer Wien New York, 2010.
• [18] I-Cheng Yeh. Design of high-performance concrete mixture using neural networks and nonlinear programming. Journal of Computing in Civil Engineering, 13(1): 36–42, 1999.