(:unkn) unknown · TIB Repositorium · 2025
Uncertainty quantification (UQ) is concerned with including and characterising uncertainties in mathematical models. Major steps include the proper description of system uncertainties, analysis and efficient quantification of uncertainties in predictions and design problems, and statistical inference on uncertain parameters starting from available measurements. Research in UQ addresses fundamental mathematical and statistical challenges, but has also wide applicability in areas such as engineering, environmental, physical and biological applications. This workshop focussed on mathematical challenges at the interface of applied mathematics, probability and statistics, numerical analysis, scientific computing and application domains such as machine learning, modelling of energy production, and bifurcations in climate models. The workshop brought together experts from those disciplines to enhance their interaction, to exchange ideas and to develop new, powerful methods for UQ.
Cited by 0Open access
Srikanta Mishra, Akhil Datta‐Gupta · Elsevier eBooks · 2017
Cited by 13
Ulrich Römer · Springer theses · 2016
Cited by 32
Hiromitsu Kumamoto, Ernest J. Henley · OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) · 2009
<P>This chapter contains sections titled: <UL> <LI> <P>Introduction</P> </LI> <LI> <P>Parametric Uncertainty</P> </LI> <LI> <P>Plant-Specific Data</P> </LI> <LI> <P>Log-Normal Distribution</P> </LI> <LI> <P>Uncertainty Propagation</P> </LI> <LI> <P>Monte Carlo Propagation</P> </LI> <LI> <P>Analytical Moment Propagation</P> </LI> <LI> <P>Discrete Probability Algebra</P> </LI> <LI> <P>Summary</P> </LI> </UL> </P> <P>This chapter contains sections titled: <UL> <LI> <P>References</P> </LI> <LI> <P>Chapter Eleven Appendices</P> </LI> <LI> <P>Problems</P> </LI> </UL> </P> ]]>
Cited by 2Open access
Cook, Douglas · ScholarsArchive (Brigham Young University) · 2021
What are the limitations of scientific models? Uncertainty quantification, Sensitivity analyses, Theory-based and statistical modeling, Systems thinking, systems analysis
Cited by 0
Behnam Sadeghi, Eric Grunsky, Vera Pawlowsky‐Glahn · Encyclopedia of earth sciences series/Encyclopedia of earth sciences · 2023
Cited by 9
Liu Chu · 2021
Cited by 0
Oliver G. Ernst, Fabio Nobile, Claudia Schillings et al. · Oberwolfach Reports · 2025
Uncertainty quantification (UQ) is concerned with including and characterising uncertainties in mathematical models. Major steps include the proper description of system uncertainties, analysis and efficient quantification of uncertainties in predictions and design problems, and statistical inference on uncertain parameters starting from available measurements. Research in UQ addresses fundamental mathematical and statistical challenges, but has also wide applicability in areas such as engineering, environmental, physical and biological applications. This workshop focussed on mathematical challenges at the interface of applied mathematics, probability and statistics, numerical analysis, scientific computing and application domains such as machine learning, modelling of energy production, and bifurcations in climate models. The workshop brought together experts from those disciplines to enhance their interaction, to exchange ideas and to develop new, powerful methods for UQ.
Cited by 0Open access
Chase, Luis · CERN Document Server (European Organization for Nuclear Research) · 2019
Cited by 0
P. S. Koutsourelakis · 2010
Cited by 0
(:unkn) unknown · TIB Repositorium · 2019
Uncertainty quantification (UQ) is concerned with including and characterising uncertainties in mathematical models. Major steps comprise proper description of system uncertainties, analysis and efficient quantification of uncertainties in predictions and design problems, and statistical inference on uncertain parameters starting from available measurements. Research in UQ addresses fundamental mathematical and statistical challenges, but has also wide applicability in areas such as engineering, environmental, physical and biological applications. This workshop focussed on mathematical challenges at the interface of applied mathematics, probability and statistics, numerical analysis, scientific computing and application domains. The workshop served to bring together experts from those disciplines in order to enhance their interaction, to exchange ideas and to develop new, powerful methods for UQ.
Cited by 0Open access
Victor E. Saouma, Mohammad Amin Hariri‐Ardebili · 2021
Cited by 2
Habib N. Najm · OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) · 2016
Abstract not provided.
Cited by 0Open access
Berger, J., Smith, Leonard A. · London School of Economics and Political Science Research Online (London School of Economics and Political Science) · 2018
AIMS AND SCOPE OF JOURNAL: The Annual Review of Statistics and Its Application informs statisticians, and users of statistics about major methodological advances and the computational tools that allow for their implementation.
Cited by 0
Jiaxin Zhang, Prafulla Kumar Choubey, Kung-Hsiang Huang et al. · ArXiv.org · 2026
Although AI agents have demonstrated impressive capabilities in long-horizon reasoning, their reliability is severely hampered by the ``Spiral of Hallucination,'' where early epistemic errors propagate irreversibly. Existing methods face a dilemma: uncertainty quantification (UQ) methods typically act as passive sensors, only diagnosing risks without addressing them, while self-reflection mechanisms suffer from continuous or aimless corrections. To bridge this gap, we propose a unified Dual-Process Agentic UQ (AUQ) framework that transforms verbalized uncertainty into active, bi-directional control signals. Our architecture comprises two complementary mechanisms: System 1 (Uncertainty-Aware Memory, UAM), which implicitly propagates verbalized confidence and semantic explanations to prevent blind decision-making; and System 2 (Uncertainty-Aware Reflection, UAR), which utilizes these explanations as rational cues to trigger targeted inference-time resolution only when necessary. This enables the agent to balance efficient execution and deep deliberation dynamically. Extensive experiments on closed-loop benchmarks and open-ended deep research tasks demonstrate that our training-free approach achieves superior performance and trajectory-level calibration. We believe this principled framework AUQ represents a significant step towards reliable agents.
Cited by 0Open access
Linus Seelinger, Anne Reinarz, Mikkel B. Lykkegaard et al. · Journal of Computational Physics · 2024
Uncertainty Quantification (UQ) is vital to safety-critical model-based analyses, but the widespread adoption of sophisticated UQ methods is limited by technical complexity. In this paper, we introduce UM-Bridge (the UQ and Modeling Bridge), a high-level abstraction and software protocol that facilitates universal interoperability of UQ software with simulation codes. It breaks down the technical complexity of advanced UQ applications and enables separation of concerns between experts. UM-Bridge democratizes UQ by allowing effective interdisciplinary collaboration, accelerating the development of advanced UQ methods, and making it easy to perform UQ analyses from prototype to High Performance Computing (HPC) scale. In addition, we present a library of ready-to-run UQ benchmark problems, all easily accessible through UM-Bridge. These benchmarks support UQ methodology research, enabling reproducible performance comparisons. We demonstrate UM-Bridge with several scientific applications, harnessing HPC resources even using UQ codes not designed with HPC support.
Cited by 3Open access
Linus Seelinger, Anne Reinarz, Mikkel B. Lykkegaard et al. · arXiv (Cornell University) · 2024
Uncertainty Quantification (UQ) is vital to safety-critical model-based analyses, but the widespread adoption of sophisticated UQ methods is limited by technical complexity. In this paper, we introduce UM-Bridge (the UQ and Modeling Bridge), a high-level abstraction and software protocol that facilitates universal interoperability of UQ software with simulation codes. It breaks down the technical complexity of advanced UQ applications and enables separation of concerns between experts. UM-Bridge democratizes UQ by allowing effective interdisciplinary collaboration, accelerating the development of advanced UQ methods, and making it easy to perform UQ analyses from prototype to High Performance Computing (HPC) scale. In addition, we present a library of ready-to-run UQ benchmark problems, all easily accessible through UM-Bridge. These benchmarks support UQ methodology research, enabling reproducible performance comparisons. We demonstrate UM-Bridge with several scientific applications, harnessing HPC resources even using UQ codes not designed with HPC support.
Cited by 0Open access
Jiaxin Zhang, Prafulla Kumar Choubey, Kung-Hsiang Huang et al. · arXiv (Cornell University) · 2026
Although AI agents have demonstrated impressive capabilities in long-horizon reasoning, their reliability is severely hampered by the ``Spiral of Hallucination,'' where early epistemic errors propagate irreversibly. Existing methods face a dilemma: uncertainty quantification (UQ) methods typically act as passive sensors, only diagnosing risks without addressing them, while self-reflection mechanisms suffer from continuous or aimless corrections. To bridge this gap, we propose a unified Dual-Process Agentic UQ (AUQ) framework that transforms verbalized uncertainty into active, bi-directional control signals. Our architecture comprises two complementary mechanisms: System 1 (Uncertainty-Aware Memory, UAM), which implicitly propagates verbalized confidence and semantic explanations to prevent blind decision-making; and System 2 (Uncertainty-Aware Reflection, UAR), which utilizes these explanations as rational cues to trigger targeted inference-time resolution only when necessary. This enables the agent to balance efficient execution and deep deliberation dynamically. Extensive experiments on closed-loop benchmarks and open-ended deep research tasks demonstrate that our training-free approach achieves superior performance and trajectory-level calibration. We believe this principled framework AUQ represents a significant step towards reliable agents.
Cited by 0Open access
Computational Structural Dynamics · 2007
We present an overview of new research efforts underway at Sandia National Laboratories to understand the sources of uncertainty and error in computational structural dynamics and other physics simulations, and to quantify their effects on predictive accuracy. In order to establish confidence in computational simulations as these simulations move further from the established experimental database, a new approach to modeling and simulation validation is needed. In particular, when simulations are used to qualify the safety and reliability of systems, we believe that validation should be based upon a comprehensive quantification of uncertainties and errors from all phases of the modeling and simulation process. Uncertainty and error quantification is a two-step process, the first step being the identification of all uncertainty and error sources in each phase of modeling and simulation. The second step is the assessment and propagation of the most significant uncertainties and errors through the phases of the modeling and simulation process to the predicted response quantities. This paper outlines the phases of modeling and simulation, the distinction between uncertainty and error, and a categorization of uncertainty and error sources in each phase of modeling and simulation. We also address the question of how uncertainties in the form or structure of the model might be assessed using multiple models. Examples from linear structural dynamics are given to illustrate these concepts.
Cited by 0
Debusschere, Bert · OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) · 2018
Abstract not provided.
Cited by 0Open access