Выходит 6 номеров в год
ISSN Печать: 2152-5080
ISSN Онлайн: 2152-5099
Indexed in
GRADIENT-BASED STOCHASTIC OPTIMIZATION METHODS IN BAYESIAN EXPERIMENTAL DESIGN
Краткое описание
Optimal experimental design (OED) seeks experiments expected to yield the most useful data for some purpose. In practical circumstances where experiments are time-consuming or resource-intensive, OED can yield enormous savings. We pursue OED for nonlinear systems from a Bayesian perspective, with the goal of choosing experiments that are optimal for parameter inference. Our objective in this context is the expected information gain in model parameters, which in general can only be estimated using Monte Carlo methods. Maximizing this objective thus becomes a stochastic optimization problem. This paper develops gradient-based stochastic optimization methods for the design of experiments on a continuous parameter space. Given a Monte Carlo estimator of expected information gain, we use infinitesimal perturbation analysis to derive gradients of this estimator.We are then able to formulate two gradient-based stochastic optimization approaches: (i) Robbins-Monro stochastic approximation, and (ii) sample average approximation combined with a deterministic quasi-Newton method. A polynomial chaos approximation of the forward model accelerates objective and gradient evaluations in both cases.We discuss the implementation of these optimization methods, then conduct an empirical comparison of their performance. To demonstrate design in a nonlinear setting with partial differential equation forward models, we use the problem of sensor placement for source inversion. Numerical results yield useful guidelines on the choice of algorithm and sample sizes, assess the impact of estimator bias, and quantify tradeoffs of computational cost versus solution quality and robustness.
-
Koutsourelakis P.S., Variational Bayesian strategies for high-dimensional, stochastic design problems, Journal of Computational Physics, 308, 2016. Crossref
-
Ryan Elizabeth G., Drovandi Christopher C., McGree James M., Pettitt Anthony N., A Review of Modern Computational Algorithms for Bayesian Optimal Design, International Statistical Review, 84, 1, 2016. Crossref
-
Alexanderian Alen, Petra Noemi, Stadler Georg, Ghattas Omar, A Fast and Scalable Method for A-Optimal Design of Experiments for Infinite-dimensional Bayesian Nonlinear Inverse Problems, SIAM Journal on Scientific Computing, 38, 1, 2016. Crossref
-
Shen Weijie, Robust Parameter Designs in Computer Experiments Using Stochastic Approximation, Technometrics, 59, 4, 2017. Crossref
-
Nath Paromita, Hu Zhen, Mahadevan Sankaran, Sensor placement for calibration of spatially varying model parameters, Journal of Computational Physics, 343, 2017. Crossref
-
Ao Dan, Hu Zhen, Mahadevan Sankaran, Design of validation experiments for life prediction models, Reliability Engineering & System Safety, 165, 2017. Crossref
-
Nath Paromita, Hu Zhen, Mahadevan Sankaran, Bayesian Calibration of Spatially Varying Model Parameters with High-Dimensional Response, 19th AIAA Non-Deterministic Approaches Conference, 2017. Crossref
-
Vohra Manav, Huan Xun, Weihs Timothy P., Knio Omar M., Design analysis for optimal calibration of diffusivity in reactive multilayers, Combustion Theory and Modelling, 21, 6, 2017. Crossref
-
Capellari Giovanni, Chatzi Eleni, Mariani Stefano, Azam Saeed Eftekhar, Optimal design of sensor networks for damage detection, Procedia Engineering, 199, 2017. Crossref
-
Tsilifis Panagiotis, Ghanem Roger G., Hajali Paris, Efficient Bayesian Experimentation Using an Expected Information Gain Lower Bound, SIAM/ASA Journal on Uncertainty Quantification, 5, 1, 2017. Crossref
-
Yu Hui, Yue Hong, Halling Peter, Robust Sampling Time Design for Biochemical Systems, IFAC-PapersOnLine, 51, 18, 2018. Crossref
-
Argyris Costas, Chowdhury Sharmistha, Zabel Volkmar, Papadimitriou Costas, Bayesian optimal sensor placement for crack identification in structures using strain measurements, Structural Control and Health Monitoring, 25, 5, 2018. Crossref
-
Alexanderian Alen, Saibaba Arvind K., Efficient D-Optimal Design of Experiments for Infinite-Dimensional Bayesian Linear Inverse Problems, SIAM Journal on Scientific Computing, 40, 5, 2018. Crossref
-
Attia Ahmed, Alexanderian Alen, Saibaba Arvind K, Goal-oriented optimal design of experiments for large-scale Bayesian linear inverse problems, Inverse Problems, 34, 9, 2018. Crossref
-
Tsilifis P., Ghanem R. G., Bayesian adaptation of chaos representations using variational inference and sampling on geodesics, Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 474, 2217, 2018. Crossref
-
Horesh Lior, Conn Andrew R., Jimenez Eduardo A., van Essen Gijs M., Reduced Space Dynamics-Based Geo-Statistical Prior Sampling for Uncertainty Quantification of End Goal Decisions, in Numerical Analysis and Optimization, 134, 2015. Crossref
-
Gharamti Mohamad E., Marzouk Youssef M., Huan Xun, Hoteit Ibrahim, A Greedy Approach for Placement of Subsurface Aquifer Wells in an Ensemble Filtering Framework, in Dynamic Data-Driven Environmental Systems Science, 8964, 2015. Crossref
-
Absi Ghina N., Mahadevan Sankaran, Simulation and Sensor Optimization for Multifidelity Dynamics Model Calibration, AIAA Journal, 58, 2, 2020. Crossref
-
Walsh Scott N., Wildey Tim M., Jakeman John D., Optimal Experimental Design Using a Consistent Bayesian Approach, ASCE-ASME J Risk and Uncert in Engrg Sys Part B Mech Engrg, 4, 1, 2018. Crossref
-
Ghosh Supriyo, Mahmoudi Mohamad, Johnson Luke, Elwany Alaa, Arroyave Raymundo, Allaire Douglas, Uncertainty analysis of microsegregation during laser powder bed fusion, Modelling and Simulation in Materials Science and Engineering, 27, 3, 2019. Crossref
-
Paulson Joel A., Martin-Casas Marc, Mesbah Ali, Optimal Bayesian experiment design for nonlinear dynamic systems with chance constraints, Journal of Process Control, 77, 2019. Crossref
-
Absi G. N., Mahadevan S., Sensor Placement for Multi-Fidelity Dynamics Model Calibration, in Model Validation and Uncertainty Quantification, Volume 3, 2020. Crossref
-
Aggarwal R., Demkowicz M. J., Marzouk Y. M., Information-Driven Experimental Design in Materials Science, in Information Science for Materials Discovery and Design, 225, 2016. Crossref
-
Sapsis Themistoklis P., Output-weighted optimal sampling for Bayesian regression and rare event statistics using few samples, Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 476, 2234, 2020. Crossref
-
Carlon André Gustavo, Dia Ben Mansour, Espath Luis, Lopez Rafael Holdorf, Tempone Raúl, Nesterov-aided stochastic gradient methods using Laplace approximation for Bayesian design optimization, Computer Methods in Applied Mechanics and Engineering, 363, 2020. Crossref
-
Ghauch Ziad G., Leveraging Adapted Polynomial Chaos Metamodels for Real-Time Bayesian Updating, Journal of Verification, Validation and Uncertainty Quantification, 4, 4, 2019. Crossref
-
Christen J. Andrés, Parker Albert E., Systematic Statistical Analysis of Microbial Data from Dilution Series, Journal of Agricultural, Biological and Environmental Statistics, 25, 3, 2020. Crossref
-
Xu Zhihang, Liao Qifeng, Gaussian Process Based Expected Information Gain Computation for Bayesian Optimal Design, Entropy, 22, 2, 2020. Crossref
-
Gao Han, Zhu Xueyu, Wang Jian-Xun, A bi-fidelity surrogate modeling approach for uncertainty propagation in three-dimensional hemodynamic simulations, Computer Methods in Applied Mechanics and Engineering, 366, 2020. Crossref
-
Beck Joakim, Mansour Dia Ben, Espath Luis, Tempone Raúl, Multilevel double loop Monte Carlo and stochastic collocation methods with importance sampling for Bayesian optimal experimental design, International Journal for Numerical Methods in Engineering, 121, 15, 2020. Crossref
-
Karimi Ahmad, Taghizadeh Leila, Heitzinger Clemens, Optimal Bayesian experimental design for electrical impedance tomography in medical imaging, Computer Methods in Applied Mechanics and Engineering, 373, 2021. Crossref
-
Jung Yongsu, Lee Ikjin, Optimal design of experiments for optimization-based model calibration using Fisher information matrix, Reliability Engineering & System Safety, 216, 2021. Crossref
-
Garoli Gabriel Y., Pilotto Rafael, Nordmann Rainer, de Castro Helio F., Identification of active magnetic bearing parameters in a rotor machine using Bayesian inference with generalized polynomial chaos expansion, Journal of the Brazilian Society of Mechanical Sciences and Engineering, 43, 12, 2021. Crossref
-
Kochunas Brendan, Huan Xun, Digital Twin Concepts with Uncertainty for Nuclear Power Applications, Energies, 14, 14, 2021. Crossref
-
Pandita Piyush, Tsilifis Panagiotis, Awalgaonkar Nimish M., Bilionis Ilias, Panchal Jitesh, Surrogate-based sequential Bayesian experimental design using non-stationary Gaussian Processes, Computer Methods in Applied Mechanics and Engineering, 385, 2021. Crossref
-
Drovandi Christopher C., Tran Minh-Ngoc, Improving the Efficiency of Fully Bayesian Optimal Design of Experiments Using Randomised Quasi-Monte Carlo, Bayesian Analysis, 13, 1, 2018. Crossref
-
Burger M, Hauptmann A, Helin T, Hyvönen N, Puska J-P, Sequentially optimized projections in x-ray imaging * , Inverse Problems, 37, 7, 2021. Crossref
-
Goda Takashi, Hironaka Tomohiko, Kitade Wataru, Foster Adam, Unbiased MLMC Stochastic Gradient-Based Optimization of Bayesian Experimental Designs, SIAM Journal on Scientific Computing, 44, 1, 2022. Crossref
-
Ashenafi Yonatan, Pandita Piyush, Ghosh Sayan, Reinforcement Learning-Based Sequential Batch-Sampling for Bayesian Optimal Experimental Design, Journal of Mechanical Design, 144, 9, 2022. Crossref
-
Alexanderian Alen, Optimal experimental design for infinite-dimensional Bayesian inverse problems governed by PDEs: a review, Inverse Problems, 37, 4, 2021. Crossref
-
Paulson Joel A., Lu Congwen, COBALT: COnstrained Bayesian optimizAtion of computationaLly expensive grey-box models exploiting derivaTive information, Computers & Chemical Engineering, 160, 2022. Crossref
-
Yu Hui, Yue Hong, Wei Xian, Su Xiaoke, A Bayesian Robust Observation Design Approach for Systems with (Large) Parametric Uncertainties, IFAC-PapersOnLine, 53, 2, 2020. Crossref
-
Ghattas Omar, Willcox Karen, Learning physics-based models from data: perspectives from inverse problems and model reduction, Acta Numerica, 30, 2021. Crossref
-
Prangle Dennis, Harbisher Sophie, Gillespie Colin S., Bayesian Experimental Design Without Posterior Calculations: An Adversarial Approach, Bayesian Analysis, -1, -1, 2022. Crossref
-
Dunbar Oliver R. A., Howland Michael F., Schneider Tapio, Stuart Andrew M., Ensemble‐Based Experimental Design for Targeting Data Acquisition to Inform Climate Models, Journal of Advances in Modeling Earth Systems, 14, 9, 2022. Crossref
-
Kreitmair Monika J., Makasis Nikolas, Menberg Kathrin, Bidarmaghz Asal, Farr Gareth J., Boon David P., Choudhary Ruchi, Bayesian parameter inference for shallow subsurface modeling using field data and impacts on geothermal planning, Data-Centric Engineering, 3, 2022. Crossref
-
Cantero‐Chinchilla Sergio, Papadimitriou Costas, Chiachío Juan, Chiachío Manuel, Koumoutsakos Petros, Fabro Adriano T., Chronopoulos Dimitrios, Robust optimal sensor configuration using the value of information, Structural Control and Health Monitoring, 29, 12, 2022. Crossref