Abstract:The Koopman operator has become an essential tool for data-driven approximation of dynamical (control) systems in recent years, e.g., via extended dynamic mode decomposition. Despite its popularity, convergence results and, in particular, error bounds are still quite scarce. In this paper, we derive probabilistic bounds for the approximation error and the prediction error depending on the number of training data points; for both ordinary and stochastic differential equations. Moreover, we extend our analysis t… Show more
“…A wealth of numerical methods for the solution of the control problem (10) exist, for instance using dynamic programming [30,31] or Monte Carlo sampling [32]. More recently, reformulations as deterministic control problems via the Koopman generator, which allow for a significant reduction of the complexity, were proposed [13,33,34]. For an application to quantum control, see [35].…”
Section: Solution Of the Control Problemmentioning
Koopman operator theory has been successfully applied to problems from various research areas such as fluid dynamics, molecular dynamics, climate science, engineering, and biology. Applications include detecting metastable or coherent sets, coarse-graining, system identification, and control. There is an intricate connection between dynamical systems driven by stochastic differential equations and quantum mechanics. In this paper, we compare the ground-state transformation and Nelson's stochastic mechanics and demonstrate how data-driven methods developed for the approximation of the Koopman operator can be used to analyze quantum physics problems. Moreover, we exploit the relationship between Schrödinger operators and stochastic control problems to show that modern data-driven methods for stochastic control can be used to solve the stationary or imaginary-time Schrödinger equation. Our findings open up a new avenue towards solving Schrödinger's equation using recently developed tools from data science. * These authors contributed equally to this work 1 A certain relationship between the basic wave mechanical equation and the Fokker equation, as well as the related statistical concepts, has probably struck anyone who is sufficiently familiar with both ideas.
“…A wealth of numerical methods for the solution of the control problem (10) exist, for instance using dynamic programming [30,31] or Monte Carlo sampling [32]. More recently, reformulations as deterministic control problems via the Koopman generator, which allow for a significant reduction of the complexity, were proposed [13,33,34]. For an application to quantum control, see [35].…”
Section: Solution Of the Control Problemmentioning
Koopman operator theory has been successfully applied to problems from various research areas such as fluid dynamics, molecular dynamics, climate science, engineering, and biology. Applications include detecting metastable or coherent sets, coarse-graining, system identification, and control. There is an intricate connection between dynamical systems driven by stochastic differential equations and quantum mechanics. In this paper, we compare the ground-state transformation and Nelson's stochastic mechanics and demonstrate how data-driven methods developed for the approximation of the Koopman operator can be used to analyze quantum physics problems. Moreover, we exploit the relationship between Schrödinger operators and stochastic control problems to show that modern data-driven methods for stochastic control can be used to solve the stationary or imaginary-time Schrödinger equation. Our findings open up a new avenue towards solving Schrödinger's equation using recently developed tools from data science. * These authors contributed equally to this work 1 A certain relationship between the basic wave mechanical equation and the Fokker equation, as well as the related statistical concepts, has probably struck anyone who is sufficiently familiar with both ideas.
“…We briefly sketch the main steps of the bilinear surrogate modeling approach presented in [33], for which a finite-data error estimate was given in [32]. Considering a control u ∈ L ∞ ([0, T ], R nc ), it turns out that by control affinity of the system, also the Koopman generators are control affine, cf.…”
Section: Bilinear Edmd-based Approximation Of Control Systemsmentioning
confidence: 99%
“…In our work [32], we provided first quantitative error bounds for approximating control systems by means of bilinear eDMD-based surrogate modeling of control systems as described in Subsection 2.2. Our results were formulated for the broad class of Stochastic Differential Equations (SDEs) from which, the ODE-dynamics (1) can be obtained as a particular case.…”
“…Constrained data-based control with guarantees. In this paper, we propose an approach based on recently-derived error estimates [32] for control-affine systems governed by ordinary (and/or stochastic) differential equations to successfully accomplish this task. To this end, we proceed as follows.…”
Section: Introduction and Outlinementioning
confidence: 99%
“…data points using [39]. 1 The respective error estimates are generalized to the control setting in [32] using ideas from [33] such that a bilinear surrogate model is obtained. This error decays with an increasing number of data points.…”
We present an approach for guaranteed constraint satisfaction by means of data-based optimal control, where the model is unknown and has to be obtained from measurement data. To this end, we utilize the Koopman framework and an eDMD-based bilinear surrogate modeling approach for control systems to show an error bound on predicted observables, i.e., functions of the state. This result is then applied to the constraints of the optimal control problem to show that satisfaction of tightened constraints in the purely data-based surrogate model implies constraint satisfaction for the original system. Keywords. Approximation error, control of constrained systems, data-based control, eDMD, finite-data error quantization, Koopman operator, MPC, nonlinear predictive control Notation: Let N 0 and R denote the natural (including zero) and the real numbers, respectively. Further, let C k ([0, T ], R), k ∈ N 0 , and L ∞ ([0, T ], R) be the spaces of k-times continuously differentiable and of Lebesgue-measurable, essentially bounded functions, respectively.
Koopman operator theory has been successfully applied to problems from various research areas such as fluid dynamics, molecular dynamics, climate science, engineering, and biology. Applications include detecting metastable or coherent sets, coarse-graining, system identification, and control. There is an intricate connection between dynamical systems driven by stochastic differential equations and quantum mechanics. In this paper, we compare the ground-state transformation and Nelson’s stochastic mechanics and demonstrate how data-driven methods developed for the approximation of the Koopman operator can be used to analyze quantum physics problems. Moreover, we exploit the relationship between Schrödinger operators and stochastic control problems to show that modern data-driven methods for stochastic control can be used to solve the stationary or imaginary-time Schrödinger equation. Our findings open up a new avenue toward solving Schrödinger’s equation using recently developed tools from data science.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.