## 2018 |

Eldred, Michael S; Geraci, Gianluca; Gorodetsky, Alex; Jakeman, John Multilevel-Multifidelity Approaches for Forward UQ in the DARPA SEQUOIA project Inproceedings 2018 AIAA Non-Deterministic Approaches Conference, AIAA SciTech Forum, Kissimmee, Florida, 2018. Abstract | Links | BibTeX | Tags: Multifidelity Methods, Polynomial Approximation, Surrogate models, Tensor decompositions @inproceedings{Eldred2018, title = {Multilevel-Multifidelity Approaches for Forward UQ in the DARPA SEQUOIA project}, author = {Michael S. Eldred and Gianluca Geraci and Alex Gorodetsky and John Jakeman}, url = {https://alexgorodetsky.com/wp-content/uploads/2018/02/main-1.pdf}, doi = {10.2514/6.2018-1179 }, year = {2018}, date = {2018-01-08}, booktitle = {2018 AIAA Non-Deterministic Approaches Conference, AIAA SciTech Forum}, address = {Kissimmee, Florida}, abstract = {Within the SEQUOIA project, funded by the DARPA EQUiPS program, we pursue algorithmic approaches that enable comprehensive design under uncertainty, through in- clusion of aleatory/parametric and epistemic/model form uncertainties within scalable for- ward/inverse UQ approaches. These statistical methods are embedded within design pro- cesses that manage computational expense through active subspace, multilevel-multifidelity, and reduced-order modeling approximations. To demonstrate these methods, we focus on the design of devices that involve multi-physics interactions in advanced aerospace vehicles. A particular problem of interest is the shape design of nozzles for advanced vehicles such as the Northrop Grumman UCAS X-47B, involving coupled aero-structural-thermal simu- lations for nozzle performance. In this paper, we explore a combination of multilevel and multifidelity forward and inverse UQ algorithms to reduce the overall computational cost of the analysis by leveraging hierarchies of model form (i.e., multifidelity hierarchies) and solution discretization (i.e., multilevel hierarchies) in order of exploit trade offs between solution accuracy and cost. In particular, we seek the most cost effective fusion of informa- tion across complex multi-dimensional modeling hierarchies. Results to date indicate the utility of multiple approaches, including methods that optimally allocate resources when estimator variance varies smoothly across levels, methods that allocate sufficient sampling density based on sparsity estimates, and methods that employ greedy multilevel refinement.}, keywords = {Multifidelity Methods, Polynomial Approximation, Surrogate models, Tensor decompositions}, pubstate = {published}, tppubtype = {inproceedings} } Within the SEQUOIA project, funded by the DARPA EQUiPS program, we pursue algorithmic approaches that enable comprehensive design under uncertainty, through in- clusion of aleatory/parametric and epistemic/model form uncertainties within scalable for- ward/inverse UQ approaches. These statistical methods are embedded within design pro- cesses that manage computational expense through active subspace, multilevel-multifidelity, and reduced-order modeling approximations. To demonstrate these methods, we focus on the design of devices that involve multi-physics interactions in advanced aerospace vehicles. A particular problem of interest is the shape design of nozzles for advanced vehicles such as the Northrop Grumman UCAS X-47B, involving coupled aero-structural-thermal simu- lations for nozzle performance. In this paper, we explore a combination of multilevel and multifidelity forward and inverse UQ algorithms to reduce the overall computational cost of the analysis by leveraging hierarchies of model form (i.e., multifidelity hierarchies) and solution discretization (i.e., multilevel hierarchies) in order of exploit trade offs between solution accuracy and cost. In particular, we seek the most cost effective fusion of informa- tion across complex multi-dimensional modeling hierarchies. Results to date indicate the utility of multiple approaches, including methods that optimally allocate resources when estimator variance varies smoothly across levels, methods that allocate sufficient sampling density based on sparsity estimates, and methods that employ greedy multilevel refinement. |

Gorodetsky, Alex; Jakeman, John Gradient-based Optimization for Regression in the Functional Tensor-Train Format Journal Article Forthcoming Arxiv 1801.00885v2 (Accepted Journal of Computational Physics), Forthcoming. Abstract | Links | BibTeX | Tags: Regression, Surrogate models, Tensor decompositions @article{Gorodetsky2018b, title = {Gradient-based Optimization for Regression in the Functional Tensor-Train Format}, author = {Alex Gorodetsky and John Jakeman}, url = {https://arxiv.org/abs/1801.00885v2}, year = {2018}, date = {2018-01-01}, journal = {Arxiv 1801.00885v2 (Accepted Journal of Computational Physics)}, abstract = {We consider the task of low-multilinear-rank functional regression, i.e., learning a low-rank parametric representation of functions from scattered real-valued data. Our first contribution is the development and analysis of an efficient gradient computation that enables gradient-based optimization procedures, including stochastic gradient descent and quasi-Newton methods, for learning the parameters of a functional tensor-train (FT). The functional tensor-train uses the tensor-train (TT) representation of low-rank arrays as an ansatz for a class of low-multilinear-rank functions. The FT is represented by a set of matrix-valued functions that contain a set of univariate functions, and the regression task is to learn the parameters of these univariate functions. Our second contribution demonstrates that using nonlinearly parameterized univariate functions, e.g., symmetric kernels with moving centers, within each core can outperform the standard approach of using a linear expansion of basis functions. Our final contributions are new rank adaptation and group-sparsity regularization procedures to minimize overfitting. We use several benchmark problems to demonstrate at least an order of magnitude lower accuracy with gradient-based optimization methods than standard alternating least squares procedures in the low-sample number regime. We also demonstrate an order of magnitude reduction in accuracy on a test problem resulting from using nonlinear parameterizations over linear parameterizations. Finally we compare regression performance with 22 other nonparametric and parametric regression methods on 10 real-world data sets. We achieve top-five accuracy for seven of the data sets and best accuracy for two of the data sets. These rankings are the best amongst parametric models and competetive with the best non-parametric methods.}, keywords = {Regression, Surrogate models, Tensor decompositions}, pubstate = {forthcoming}, tppubtype = {article} } We consider the task of low-multilinear-rank functional regression, i.e., learning a low-rank parametric representation of functions from scattered real-valued data. Our first contribution is the development and analysis of an efficient gradient computation that enables gradient-based optimization procedures, including stochastic gradient descent and quasi-Newton methods, for learning the parameters of a functional tensor-train (FT). The functional tensor-train uses the tensor-train (TT) representation of low-rank arrays as an ansatz for a class of low-multilinear-rank functions. The FT is represented by a set of matrix-valued functions that contain a set of univariate functions, and the regression task is to learn the parameters of these univariate functions. Our second contribution demonstrates that using nonlinearly parameterized univariate functions, e.g., symmetric kernels with moving centers, within each core can outperform the standard approach of using a linear expansion of basis functions. Our final contributions are new rank adaptation and group-sparsity regularization procedures to minimize overfitting. We use several benchmark problems to demonstrate at least an order of magnitude lower accuracy with gradient-based optimization methods than standard alternating least squares procedures in the low-sample number regime. We also demonstrate an order of magnitude reduction in accuracy on a test problem resulting from using nonlinear parameterizations over linear parameterizations. Finally we compare regression performance with 22 other nonparametric and parametric regression methods on 10 real-world data sets. We achieve top-five accuracy for seven of the data sets and best accuracy for two of the data sets. These rankings are the best amongst parametric models and competetive with the best non-parametric methods. |

## 2017 |

Gorodetsky, Alex Massachusetts Institute of Technology, 2017. Abstract | Links | BibTeX | Tags: Dynamic Programming, Kalman Filtering, Stochastic Optimal Control, Surrogate models, Tensor decompositions @phdthesis{Gorodetsky2017b, title = {Continuous low-rank tensor decompositions, with applications to stochastic optimal control and data assimilation}, author = {Alex Gorodetsky}, url = {http://hdl.handle.net/1721.1/108918}, year = {2017}, date = {2017-02-02}, address = {Cambridge, MA}, school = {Massachusetts Institute of Technology}, abstract = {Optimal decision making under uncertainty is critical for control and optimization of complex systems. However, many techniques for solving problems such as stochastic optimal control and data assimilation encounter the curse of dimensionality when too many state variables are involved. In this thesis, we propose a framework for computing with high-dimensional functions that mitigates this exponential growth in complexity for problems with separable structure. Our framework tightly integrates two emerging areas: tensor decompositions and continuous computation. Tensor decompositions are able to effectively compress and operate with low-rank multidimensional arrays. Continuous computation is a paradigm for computing with functions instead of arrays, and it is best realized by Chebfun, a MATLAB package for computing with functions of up to three dimensions. Continuous computation provides a natural framework for building numerical algorithms that effectively, naturally, and automatically adapt to problem structure. The first part of this thesis describes a compressed continuous computation framework centered around a continuous analogue to the (discrete) tensor-train decomposition called the function-train decomposition. Computation with the function-train requires continuous matrix factorizations and continuous numerical linear algebra. Continuous analogues are presented for performing cross approximation; rounding; multilinear algebra operations such as addition, multiplication, integration, and differentiation; and continuous, rank-revealing, alternating least squares. Advantages of the function-train over the tensor-train include the ability to adaptively approximate functions and the ability to compute with functions that are parameterized differently. For example, while elementwise multiplication between tensors of different sizes is undefined, functions in FT format can be readily multiplied together. Next, we develop compressed versions of value iteration, policy iteration, and multilevel algorithms for solving dynamic programming problems arising in stochastic optimal control. These techniques enable computing global solutions to a broader set of problems, for example those with non-affine control inputs, than previously possible. Examples are presented for motion planning with robotic systems that have up to seven states. Finally, we use the FT to extend integration-based Gaussian filtering to larger state spaces than previously considered. Examples are presented for dynamical systems with up to twenty states.}, keywords = {Dynamic Programming, Kalman Filtering, Stochastic Optimal Control, Surrogate models, Tensor decompositions}, pubstate = {published}, tppubtype = {phdthesis} } Optimal decision making under uncertainty is critical for control and optimization of complex systems. However, many techniques for solving problems such as stochastic optimal control and data assimilation encounter the curse of dimensionality when too many state variables are involved. In this thesis, we propose a framework for computing with high-dimensional functions that mitigates this exponential growth in complexity for problems with separable structure. Our framework tightly integrates two emerging areas: tensor decompositions and continuous computation. Tensor decompositions are able to effectively compress and operate with low-rank multidimensional arrays. Continuous computation is a paradigm for computing with functions instead of arrays, and it is best realized by Chebfun, a MATLAB package for computing with functions of up to three dimensions. Continuous computation provides a natural framework for building numerical algorithms that effectively, naturally, and automatically adapt to problem structure. The first part of this thesis describes a compressed continuous computation framework centered around a continuous analogue to the (discrete) tensor-train decomposition called the function-train decomposition. Computation with the function-train requires continuous matrix factorizations and continuous numerical linear algebra. Continuous analogues are presented for performing cross approximation; rounding; multilinear algebra operations such as addition, multiplication, integration, and differentiation; and continuous, rank-revealing, alternating least squares. Advantages of the function-train over the tensor-train include the ability to adaptively approximate functions and the ability to compute with functions that are parameterized differently. For example, while elementwise multiplication between tensors of different sizes is undefined, functions in FT format can be readily multiplied together. Next, we develop compressed versions of value iteration, policy iteration, and multilevel algorithms for solving dynamic programming problems arising in stochastic optimal control. These techniques enable computing global solutions to a broader set of problems, for example those with non-affine control inputs, than previously possible. Examples are presented for motion planning with robotic systems that have up to seven states. Finally, we use the FT to extend integration-based Gaussian filtering to larger state spaces than previously considered. Examples are presented for dynamical systems with up to twenty states. |

## 2014 |

Gorodetsky, Alex; Marzouk, Youssef Efficient Localization of Discontinuities in Complex Computational Simulations Journal Article SIAM Journal on Scientific Computing, 36 (6), pp. A2584-A2610, 2014, ISSN: 1095-7197. Abstract | Links | BibTeX | Tags: Active Learning, Support Vector Machines, Surrogate models @article{Gorodetsky2014, title = {Efficient Localization of Discontinuities in Complex Computational Simulations }, author = {Alex Gorodetsky and Youssef Marzouk}, url = {https://alexgorodetsky.com/wp-content/uploads/2018/02/discontinuity_detection.pdf}, doi = {10.1137/140953137}, issn = {1095-7197}, year = {2014}, date = {2014-11-04}, journal = {SIAM Journal on Scientific Computing}, volume = {36}, number = {6}, pages = {A2584-A2610}, abstract = {Surrogate models for computational simulations are input-output approximations that allow computationally intensive analyses, such as uncertainty propagation and inference, to be performed efficiently. When a simulation output does not depend smoothly on its inputs, the error and convergence rate of many approximation methods deteriorate substantially. This paper details a method for efficiently localizing discontinuities in the input parameter domain, so that the model output can be approximated as a piecewise smooth function. The approach comprises an initialization phase, which uses polynomial annihilation to assign function values to different regions and thus seed an automated labeling procedure, followed by a refinement phase that adaptively updates a kernel support vector machine representation of the separating surface via active learning. The overall approach avoids structured grids and exploits any available simplicity in the geometry of the separating surface, thus reducing the number of model evaluations required to localize the discontinuity. The method is illustrated on examples of up to eleven dimensions, including algebraic models and ODE/PDE systems, and demonstrates improved scaling and efficiency over other discontinuity localization approaches.}, keywords = {Active Learning, Support Vector Machines, Surrogate models}, pubstate = {published}, tppubtype = {article} } Surrogate models for computational simulations are input-output approximations that allow computationally intensive analyses, such as uncertainty propagation and inference, to be performed efficiently. When a simulation output does not depend smoothly on its inputs, the error and convergence rate of many approximation methods deteriorate substantially. This paper details a method for efficiently localizing discontinuities in the input parameter domain, so that the model output can be approximated as a piecewise smooth function. The approach comprises an initialization phase, which uses polynomial annihilation to assign function values to different regions and thus seed an automated labeling procedure, followed by a refinement phase that adaptively updates a kernel support vector machine representation of the separating surface via active learning. The overall approach avoids structured grids and exploits any available simplicity in the geometry of the separating surface, thus reducing the number of model evaluations required to localize the discontinuity. The method is illustrated on examples of up to eleven dimensions, including algebraic models and ODE/PDE systems, and demonstrates improved scaling and efficiency over other discontinuity localization approaches. |