TY - JOUR A1 - Acevedo, Walter A1 - De Wiljes, Jana A1 - Reich, Sebastian T1 - Second-order accurate ensemble transform particle filters JF - SIAM journal on scientific computing N2 - Particle filters (also called sequential Monte Carlo methods) are widely used for state and parameter estimation problems in the context of nonlinear evolution equations. The recently proposed ensemble transform particle filter (ETPF) [S. Reich, SIAM T. Sci. Comput., 35, (2013), pp. A2013-A2014[ replaces the resampling step of a standard particle filter by a linear transformation which allows for a hybridization of particle filters with ensemble Kalman filters and renders the resulting hybrid filters applicable to spatially extended systems. However, the linear transformation step is computationally expensive and leads to an underestimation of the ensemble spread for small and moderate ensemble sizes. Here we address both of these shortcomings by developing second order accurate extensions of the ETPF. These extensions allow one in particular to replace the exact solution of a linear transport problem by its Sinkhorn approximation. It is also demonstrated that the nonlinear ensemble transform filter arises as a special case of our general framework. We illustrate the performance of the second-order accurate filters for the chaotic Lorenz-63 and Lorenz-96 models and a dynamic scene-viewing model. The numerical results for the Lorenz-63 and Lorenz-96 models demonstrate that significant accuracy improvements can be achieved in comparison to a standard ensemble Kalman filter and the ETPF for small to moderate ensemble sizes. The numerical results for the scene-viewing model reveal, on the other hand, that second-order corrections can lead to statistically inconsistent samples from the posterior parameter distribution. KW - Bayesian inference KW - data assimilation KW - particle filter KW - ensemble Kalman filter KW - Sinkhorn approximation Y1 - 2017 U6 - https://doi.org/10.1137/16M1095184 SN - 1064-8275 SN - 1095-7197 SN - 2168-3417 VL - 39 IS - 5 SP - A1834 EP - A1850 PB - Society for Industrial and Applied Mathematics CY - Philadelphia ER - TY - JOUR A1 - Pathiraja, Sahani Darschika A1 - Reich, Sebastian A1 - Stannat, Wilhelm T1 - McKean-Vlasov SDEs in nonlinear filtering JF - SIAM journal on control and optimization : a publication of the Society for Industrial and Applied Mathematics N2 - Various particle filters have been proposed over the last couple of decades with the common feature that the update step is governed by a type of control law. This feature makes them an attractive alternative to traditional sequential Monte Carlo which scales poorly with the state dimension due to weight degeneracy. This article proposes a unifying framework that allows us to systematically derive the McKean-Vlasov representations of these filters for the discrete time and continuous time observation case, taking inspiration from the smooth approximation of the data considered in [D. Crisan and J. Xiong, Stochastics, 82 (2010), pp. 53-68; J. M. Clark and D. Crisan, Probab. Theory Related Fields, 133 (2005), pp. 43-56]. We consider three filters that have been proposed in the literature and use this framework to derive Ito representations of their limiting forms as the approximation parameter delta -> 0. All filters require the solution of a Poisson equation defined on R-d, for which existence and uniqueness of solutions can be a nontrivial issue. We additionally establish conditions on the signal-observation system that ensures well-posedness of the weighted Poisson equation arising in one of the filters. KW - data assimilation KW - feedback particle filter KW - Poincare inequality KW - well-posedness KW - nonlinear filtering KW - McKean-Vlasov KW - mean-field equations Y1 - 2022 U6 - https://doi.org/10.1137/20M1355197 SN - 0363-0129 SN - 1095-7138 VL - 59 IS - 6 SP - 4188 EP - 4215 PB - Society for Industrial and Applied Mathematics CY - Philadelphia ER - TY - JOUR A1 - de Wiljes, Jana A1 - Reich, Sebastian A1 - Stannat, Wilhelm T1 - Long-Time stability and accuracy of the ensemble Kalman-Bucy Filter for fully observed processes and small measurement noise JF - SIAM Journal on Applied Dynamical Systems N2 - The ensemble Kalman filter has become a popular data assimilation technique in the geosciences. However, little is known theoretically about its long term stability and accuracy. In this paper, we investigate the behavior of an ensemble Kalman-Bucy filter applied to continuous-time filtering problems. We derive mean field limiting equations as the ensemble size goes to infinity as well as uniform-in-time accuracy and stability results for finite ensemble sizes. The later results require that the process is fully observed and that the measurement noise is small. We also demonstrate that our ensemble Kalman-Bucy filter is consistent with the classic Kalman-Bucy filter for linear systems and Gaussian processes. We finally verify our theoretical findings for the Lorenz-63 system. KW - data assimilation KW - Kalman Bucy filter KW - ensemble Kalman filter KW - stability KW - accuracy KW - asymptotic behavior Y1 - 2018 U6 - https://doi.org/10.1137/17M1119056 SN - 1536-0040 VL - 17 IS - 2 SP - 1152 EP - 1181 PB - Society for Industrial and Applied Mathematics CY - Philadelphia ER - TY - JOUR A1 - de Wiljes, Jana A1 - Pathiraja, Sahani Darschika A1 - Reich, Sebastian T1 - Ensemble transform algorithms for nonlinear smoothing problems JF - SIAM journal on scientific computing N2 - Several numerical tools designed to overcome the challenges of smoothing in a non-linear and non-Gaussian setting are investigated for a class of particle smoothers. The considered family of smoothers is induced by the class of linear ensemble transform filters which contains classical filters such as the stochastic ensemble Kalman filter, the ensemble square root filter, and the recently introduced nonlinear ensemble transform filter. Further the ensemble transform particle smoother is introduced and particularly highlighted as it is consistent in the particle limit and does not require assumptions with respect to the family of the posterior distribution. The linear update pattern of the considered class of linear ensemble transform smoothers allows one to implement important supplementary techniques such as adaptive spread corrections, hybrid formulations, and localization in order to facilitate their application to complex estimation problems. These additional features are derived and numerically investigated for a sequence of increasingly challenging test problems. KW - data assimilation KW - smoother KW - localization KW - optimal transport KW - adaptive KW - spread correction Y1 - 2019 U6 - https://doi.org/10.1137/19M1239544 SN - 1064-8275 SN - 1095-7197 VL - 42 IS - 1 SP - A87 EP - A114 PB - Society for Industrial and Applied Mathematics CY - Philadelphia ER - TY - JOUR A1 - Pathiraja, Sahani Darschika A1 - Moradkhani, H. A1 - Marshall, L. A1 - Sharma, Ashish A1 - Geenens, G. T1 - Data-driven model uncertainty estimation in hydrologic data assimilation JF - Water resources research : WRR / American Geophysical Union N2 - The increasing availability of earth observations necessitates mathematical methods to optimally combine such data with hydrologic models. Several algorithms exist for such purposes, under the umbrella of data assimilation (DA). However, DA methods are often applied in a suboptimal fashion for complex real-world problems, due largely to several practical implementation issues. One such issue is error characterization, which is known to be critical for a successful assimilation. Mischaracterized errors lead to suboptimal forecasts, and in the worst case, to degraded estimates even compared to the no assimilation case. Model uncertainty characterization has received little attention relative to other aspects of DA science. Traditional methods rely on subjective, ad hoc tuning factors or parametric distribution assumptions that may not always be applicable. We propose a novel data-driven approach (named SDMU) to model uncertainty characterization for DA studies where (1) the system states are partially observed and (2) minimal prior knowledge of the model error processes is available, except that the errors display state dependence. It includes an approach for estimating the uncertainty in hidden model states, with the end goal of improving predictions of observed variables. The SDMU is therefore suited to DA studies where the observed variables are of primary interest. Its efficacy is demonstrated through a synthetic case study with low-dimensional chaotic dynamics and a real hydrologic experiment for one-day-ahead streamflow forecasting. In both experiments, the proposed method leads to substantial improvements in the hidden states and observed system outputs over a standard method involving perturbation with Gaussian noise. KW - data assimilation KW - model error KW - uncertainty quantification KW - particle filter KW - nonparametric statistics Y1 - 2018 U6 - https://doi.org/10.1002/2018WR022627 SN - 0043-1397 SN - 1944-7973 VL - 54 IS - 2 SP - 1252 EP - 1280 PB - American Geophysical Union CY - Washington ER - TY - THES A1 - Maier, Corinna T1 - Bayesian data assimilation and reinforcement learning for model-informed precision dosing in oncology T1 - Bayes’sche Datenassimilation und Reinforcement Learning für die modellinformierte Präzisionsdosierung in der Onkologie N2 - While patients are known to respond differently to drug therapies, current clinical practice often still follows a standardized dosage regimen for all patients. For drugs with a narrow range of both effective and safe concentrations, this approach may lead to a high incidence of adverse events or subtherapeutic dosing in the presence of high patient variability. Model-informedprecision dosing (MIPD) is a quantitative approach towards dose individualization based on mathematical modeling of dose-response relationships integrating therapeutic drug/biomarker monitoring (TDM) data. MIPD may considerably improve the efficacy and safety of many drug therapies. Current MIPD approaches, however, rely either on pre-calculated dosing tables or on simple point predictions of the therapy outcome. These approaches lack a quantification of uncertainties and the ability to account for effects that are delayed. In addition, the underlying models are not improved while applied to patient data. Therefore, current approaches are not well suited for informed clinical decision-making based on a differentiated understanding of the individually predicted therapy outcome. The objective of this thesis is to develop mathematical approaches for MIPD, which (i) provide efficient fully Bayesian forecasting of the individual therapy outcome including associated uncertainties, (ii) integrate Markov decision processes via reinforcement learning (RL) for a comprehensive decision framework for dose individualization, (iii) allow for continuous learning across patients and hospitals. Cytotoxic anticancer chemotherapy with its major dose-limiting toxicity, neutropenia, serves as a therapeutically relevant application example. For more comprehensive therapy forecasting, we apply Bayesian data assimilation (DA) approaches, integrating patient-specific TDM data into mathematical models of chemotherapy-induced neutropenia that build on prior population analyses. The value of uncertainty quantification is demonstrated as it allows reliable computation of the patient-specific probabilities of relevant clinical quantities, e.g., the neutropenia grade. In view of novel home monitoring devices that increase the amount of TDM data available, the data processing of sequential DA methods proves to be more efficient and facilitates handling of the variability between dosing events. By transferring concepts from DA and RL we develop novel approaches for MIPD. While DA-guided dosing integrates individualized uncertainties into dose selection, RL-guided dosing provides a framework to consider delayed effects of dose selections. The combined DA-RL approach takes into account both aspects simultaneously and thus represents a holistic approach towards MIPD. Additionally, we show that RL can be used to gain insights into important patient characteristics for dose selection. The novel dosing strategies substantially reduce the occurrence of both subtherapeutic and life-threatening neutropenia grades in a simulation study based on a recent clinical study (CEPAC-TDM trial) compared to currently used MIPD approaches. If MIPD is to be implemented in routine clinical practice, a certain model bias with respect to the underlying model is inevitable, as the models are typically based on data from comparably small clinical trials that reflect only to a limited extent the diversity in real-world patient populations. We propose a sequential hierarchical Bayesian inference framework that enables continuous cross-patient learning to learn the underlying model parameters of the target patient population. It is important to note that the approach only requires summary information of the individual patient data to update the model. This separation of the individual inference from population inference enables implementation across different centers of care. The proposed approaches substantially improve current MIPD approaches, taking into account new trends in health care and aspects of practical applicability. They enable progress towards more informed clinical decision-making, ultimately increasing patient benefits beyond the current practice. N2 - Obwohl Patienten sehr unterschiedlich auf medikamentöse Therapien ansprechen, werden in der klinischen Praxis häufig noch standardisierte Dosierungsschemata angewendet. Bei Arzneimitteln mit engen therapeutischen Fenstern zwischen minimal wirksamen und toxischen Konzentrationen kann dieser Ansatz bei hoher interindividueller Variabilität zu häufigem Auftreten von Toxizitäten oder subtherapeutischen Konzentrationen führen. Die modellinformierte Präzisionsdosierung (MIPD) ist ein quantitativer Ansatz zur Dosisindividualisierung, der auf der mathematischen Modellierung von Dosis-Wirkungs-Beziehungen beruht und Daten aus dem therapeutischen Drug/Biomarker-Monitoring (TDM) einbezieht. Die derzeitigen MIPD-Ansätze verwenden entweder Dosierungstabellen oder einfache Punkt-Vorhersagen des Therapieverlaufs. Diesen Ansätzen fehlt eine Quantifizierung der Unsicherheiten, verzögerte Effekte werden nicht berücksichtigt und die zugrunde liegenden Modelle werden im Laufe der Anwendung nicht verbessert. Daher sind die derzeitigen Ansätze nicht ideal für eine fundierte klinische Entscheidungsfindung auf Grundlage eines differenzierten Verständnisses des individuell vorhergesagten Therapieverlaufs. Das Ziel dieser Arbeit ist es, mathematische Ansätze für das MIPD zu entwickeln, die (i) eine effiziente, vollständig Bayes’sche Vorhersage des individuellen Therapieverlaufs einschließlich der damit verbundenen Unsicherheiten ermöglichen, (ii) Markov-Entscheidungsprozesse mittels Reinforcement Learning (RL) in einen umfassenden Entscheidungsrahmen zur Dosisindividualisierung integrieren, und (iii) ein kontinuierliches Lernen zwischen Patienten erlauben. Die antineoplastische Chemotherapie mit ihrer wichtigen dosislimitierenden Toxizität, der Neutropenie, dient als therapeutisch relevantes Anwendungsbeispiel. Für eine umfassendere Therapievorhersage wenden wir Bayes’sche Datenassimilationsansätze (DA) an, um TDM-Daten in mathematische Modelle der Chemotherapie-induzierten Neutropenie zu integrieren. Wir zeigen, dass die Quantifizierung von Unsicherheiten einen großen Mehrwert bietet, da sie eine zuverlässige Berechnung der Wahrscheinlichkeiten relevanter klinischer Größen, z.B. des Neutropeniegrades, ermöglicht. Im Hinblick auf neue Home-Monitoring-Geräte, die die Anzahl der verfügbaren TDM-Daten erhöhen, erweisen sich sequenzielle DA-Methoden als effizienter und erleichtern den Umgang mit der Unsicherheit zwischen Dosierungsereignissen. Basierend auf Konzepten aus DA und RL, entwickeln wir neue Ansätze für MIPD. Während die DA-geleitete Dosierung individualisierte Unsicherheiten in die Dosisauswahl integriert, berücksichtigt die RL-geleitete Dosierung verzögerte Effekte der Dosisauswahl. Der kombinierte DA-RL-Ansatz vereint beide Aspekte und stellt somit einen ganzheitlichen Ansatz für MIPD dar. Zusätzlich zeigen wir, dass RL Informationen über die für die Dosisauswahl relevanten Patientencharakteristika liefert. Der Vergleich zu derzeit verwendeten MIPD Ansätzen in einer auf einer klinischen Studie (CEPAC-TDM-Studie) basierenden Simulationsstudie zeigt, dass die entwickelten Dosierungsstrategien das Auftreten subtherapeutischer Konzentrationen sowie lebensbedrohlicher Neutropenien drastisch reduzieren. Wird MIPD in der klinischen Routine eingesetzt, ist eine gewisse Modellverzerrung unvermeidlich. Die Modelle basieren in der Regel auf Daten aus vergleichsweise kleinen klinischen Studien, die die Heterogenität realer Patientenpopulationen nur begrenzt widerspiegeln. Wir schlagen einen sequenziellen hierarchischen Bayes’schen Inferenzrahmen vor, der ein kontinuierliches patientenübergreifendes Lernen ermöglicht, um die zugrunde liegenden Modellparameter der Ziel-Patientenpopulation zu erlernen. Zur Aktualisierung des Modells erfordert dieser Ansatz lediglich zusammenfassende Informationen der individuellen Patientendaten, was eine Umsetzung über verschiedene Versorgungszentren hinweg erlaubt. Die vorgeschlagenen Ansätze verbessern die derzeitigen MIPD-Ansätze erheblich, wobei neue Trends in der Gesundheitsversorgung und Aspekte der praktischen Anwendbarkeit berücksichtigt werden. Damit stellen sie einen Fortschritt in Richtung einer fundierteren klinischen Entscheidungsfindung dar. KW - data assimilation KW - Datenassimilation KW - reinforcement learning KW - model-informed precision dosing KW - pharmacometrics KW - oncology KW - modellinformierte Präzisionsdosierung KW - Onkologie KW - Pharmakometrie KW - Reinforcement Learning Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-515870 ER - TY - JOUR A1 - Hastermann, Gottfried A1 - Reinhardt, Maria A1 - Klein, Rupert A1 - Reich, Sebastian T1 - Balanced data assimilation for highly oscillatory mechanical systems JF - Communications in applied mathematics and computational science : CAMCoS N2 - Data assimilation algorithms are used to estimate the states of a dynamical system using partial and noisy observations. The ensemble Kalman filter has become a popular data assimilation scheme due to its simplicity and robustness for a wide range of application areas. Nevertheless, this filter also has limitations due to its inherent assumptions of Gaussianity and linearity, which can manifest themselves in the form of dynamically inconsistent state estimates. This issue is investigated here for balanced, slowly evolving solutions to highly oscillatory Hamiltonian systems which are prototypical for applications in numerical weather prediction. It is demonstrated that the standard ensemble Kalman filter can lead to state estimates that do not satisfy the pertinent balance relations and ultimately lead to filter divergence. Two remedies are proposed, one in terms of blended asymptotically consistent time-stepping schemes, and one in terms of minimization-based postprocessing methods. The effects of these modifications to the standard ensemble Kalman filter are discussed and demonstrated numerically for balanced motions of two prototypical Hamiltonian reference systems. KW - data assimilation KW - ensemble Kalman filter KW - balanced dynamics KW - highly KW - oscillatory systems KW - Hamiltonian dynamics KW - geophysics Y1 - 2021 U6 - https://doi.org/10.2140/camcos.2021.16.119 SN - 1559-3940 SN - 2157-5452 VL - 16 IS - 1 SP - 119 EP - 154 PB - Mathematical Sciences Publishers CY - Berkeley ER - TY - GEN A1 - Wiljes, Jana de A1 - Tong, Xin T. T1 - Analysis of a localised nonlinear ensemble Kalman Bucy filter with complete and accurate observations T2 - Postprints der Universität Potsdam : Mathematisch-Naturwissenschaftliche Reihe N2 - Concurrent observation technologies have made high-precision real-time data available in large quantities. Data assimilation (DA) is concerned with how to combine this data with physical models to produce accurate predictions. For spatial-temporal models, the ensemble Kalman filter with proper localisation techniques is considered to be a state-of-the-art DA methodology. This article proposes and investigates a localised ensemble Kalman Bucy filter for nonlinear models with short-range interactions. We derive dimension-independent and component-wise error bounds and show the long time path-wise error only has logarithmic dependence on the time range. The theoretical results are verified through some simple numerical tests. T3 - Zweitveröffentlichungen der Universität Potsdam : Mathematisch-Naturwissenschaftliche Reihe - 1221 KW - data assimilation KW - stability and accuracy KW - dimension independent bound KW - localisation KW - high dimensional KW - filter KW - nonlinear Y1 - 2022 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-540417 SN - 1866-8372 VL - 33 IS - 9 SP - 4752 EP - 4782 PB - IOP Publ. CY - Bristol ER - TY - JOUR A1 - Wiljes, Jana de A1 - Tong, Xin T. T1 - Analysis of a localised nonlinear ensemble Kalman Bucy filter with complete and accurate observations JF - Nonlinearity N2 - Concurrent observation technologies have made high-precision real-time data available in large quantities. Data assimilation (DA) is concerned with how to combine this data with physical models to produce accurate predictions. For spatial-temporal models, the ensemble Kalman filter with proper localisation techniques is considered to be a state-of-the-art DA methodology. This article proposes and investigates a localised ensemble Kalman Bucy filter for nonlinear models with short-range interactions. We derive dimension-independent and component-wise error bounds and show the long time path-wise error only has logarithmic dependence on the time range. The theoretical results are verified through some simple numerical tests. KW - data assimilation KW - stability and accuracy KW - dimension independent bound KW - localisation KW - high dimensional KW - filter KW - nonlinear Y1 - 2020 U6 - https://doi.org/10.1088/1361-6544/ab8d14 SN - 0951-7715 SN - 1361-6544 VL - 33 IS - 9 SP - 4752 EP - 4782 PB - IOP Publ. CY - Bristol ER - TY - JOUR A1 - Reich, Sebastian T1 - A Gaussian-mixture ensemble transform filter JF - Quarterly journal of the Royal Meteorological Society N2 - We generalize the popular ensemble Kalman filter to an ensemble transform filter, in which the prior distribution can take the form of a Gaussian mixture or a Gaussian kernel density estimator. The design of the filter is based on a continuous formulation of the Bayesian filter analysis step. We call the new filter algorithm the ensemble Gaussian-mixture filter (EGMF). The EGMF is implemented for three simple test problems (Brownian dynamics in one dimension, Langevin dynamics in two dimensions and the three-dimensional Lorenz-63 model). It is demonstrated that the EGMF is capable of tracking systems with non-Gaussian uni- and multimodal ensemble distributions. KW - data assimilation KW - ensemble Kalman filter KW - nonlinear filtering KW - Gaussian mixtures KW - Gaussian kernel estimators Y1 - 2012 U6 - https://doi.org/10.1002/qj.898 SN - 0035-9009 VL - 138 IS - 662 SP - 222 EP - 233 PB - Wiley-Blackwell CY - Malden ER -