Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 32 for “"operator learning"”.

  1. Operator learning in the overparameterized regime

    Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2025-05-01

    uiuc Repository record for Operator learning in the overparameterized regime (opens in a new tab)

  2. ADAPTIVE NEURAL NETWORK FOR DISCRETIZATION INVARIANT OPERATOR LEARNING

    Discretization invariant learning aims at learning in the infinite-dimensional function spaces with the capacity to process heterogeneous discrete representations of functions as inputs and/or outputs of a learning model. This paper proposes a novel deep learning framework based on integral …

    nus Repository record for ADAPTIVE NEURAL NETWORK FOR DISCRETIZATION INVARIANT OPERATOR LEARNING (opens in a new tab) Download file for ADAPTIVE NEURAL NETWORK FOR DISCRETIZATION INVARIANT OPERATOR LEARNING (opens in a new tab)

  3. Computer Vision and Operator Learning for Optimizing Neurosurgical Interventions

    … was created to train and benchmark deep learning models. These models localize injury sites and delineate anatomical structures with high precision. Their performance is further evaluated on human spinal cord ultrasound data to assess clinical generalization, demonstrating the potential …

    jhu Repository record for Computer Vision and Operator Learning for Optimizing Neurosurgical Interventions (opens in a new tab) Embargoed until 2027-12-01

  4. Physics-informed operator learning for solving ODE/PDEs with time-dependent forcing term

    Η τεράστια αύξηση της διαθέσιμης υπολογιστικής ισχύος έχει καταστήσει δυνατή την προσομοίωση διαφόρων προβλημάτων φυσικής επιλύοντας τις διαφορικές εξισώσεις τους γρηγορότερα. Ωστόσο, παρόλες αυτές τις τεχνολογικές εξελίξεις, η πολυπλοκότητα της όλης διαδικασίας παραμένει πρόκληση για τον εκάστοτε …

    athens Repository record for Physics-informed operator learning for solving ODE/PDEs with time-dependent forcing term (opens in a new tab)

  5. Machine Learning in Function Spaces

    Operator learning is an emerging area of machine learning which aims to learn mappings (operators) between functions from data. Many physical systems can be mathematically formulated as giving a relationship between functional data, hence operator learning has the potential to be a transformative …

    penn Repository record for Machine Learning in Function Spaces (opens in a new tab)

  6. Physics-Informed and Data-Driven Deep Learning Approaches in Scientific Computation

    The intersection of deep learning and scientific computation presents transformative opportunities to accelerate the modeling, simulation, and analysis of complex physical systems. This thesis details research projects with a focus on physics-informed and data-driven deep learning approaches in …

    tamu Repository record for Physics-Informed and Data-Driven Deep Learning Approaches in Scientific Computation (opens in a new tab)

  7. Physics-Informed and Data-Driven Deep Learning Approaches in Scientific Computation

    The intersection of deep learning and scientific computation presents transformative opportunities to accelerate the modeling, simulation, and analysis of complex physical systems. This thesis details research projects with a focus on physics-informed and data-driven deep learning approaches in …

    tamu Repository record for Physics-Informed and Data-Driven Deep Learning Approaches in Scientific Computation (opens in a new tab)

  8. Physics-Informed and Data-Driven Deep Learning Approaches in Scientific Computation

    The intersection of deep learning and scientific computation presents transformative opportunities to accelerate the modeling, simulation, and analysis of complex physical systems. This thesis details research projects with a focus on physics-informed and data-driven deep learning approaches in …

    tamu Repository record for Physics-Informed and Data-Driven Deep Learning Approaches in Scientific Computation (opens in a new tab)

  9. SOME COMPARISONS OF NEURAL NETWORK ARCHITECTURES FOR SCIENTIFIC MACHINE LEARNING

    … for approximating solutions to and solution operators for a handful of elementary 1D partial differential equations. Specifically, we examine whether residual layers offer any benefits over fully connected layers in the context of physics-informed machine learning, finding that the two …

    nps Repository record for SOME COMPARISONS OF NEURAL NETWORK ARCHITECTURES FOR SCIENTIFIC MACHINE LEARNING (opens in a new tab)

  10. Enhancing Magnetic Endoscope Manoeuvrability Using Tip-Growing and Developable Roller Mechanisms for Deep Navigation and Virtual Biopsies

    … discomfort, sedation requirements, and steep operator learning curves. The global push toward automation and patient-centric technologies has fueled interest in robotic solutions that can enhance the reach, dexterity, and diagnostic capability of endoscopic tools. This thesis is motivated by …

    whiterose Download PDF for Enhancing Magnetic Endoscope Manoeuvrability Using Tip-Growing and Developable Roller Mechanisms for Deep Navigation and Virtual Biopsies (opens in a new tab)

  11. Improving the control of two-mode flexible systems with input shaping

    … presented to verify the new algorithm. Two human operator studies are presented to demonstrate the effectiveness of input shaping when used in conjunction with distributed crane payloads, such as cargo containers. One study investigates the improvement in efficiency when using input shaping and …

    gatech Repository record for Improving the control of two-mode flexible systems with input shaping (opens in a new tab)

  12. Generative and variational approaches to scientific machine learning, inverse problems and PDEs

    … Chapter 2 develops a kernel-based framework for learning differential equations from data, together with quantitative worst-case error bounds. The state and the unknown governing operator are both represented in reproducing kernel Hilbert spaces, with the differential equation enforced via …

    washington Repository record for Generative and variational approaches to scientific machine learning, inverse problems and PDEs (opens in a new tab)

  13. Learning seismic wave propagation with Fourier Neural Operators: Advantages, limitations, and Physics-Informed implementation

    … are needed. In this context, Fourier Neural Operators (FNOs) have become a promising data-driven alternative for learning solution operators of partial differential equations. Their ability to approximate mappings between function spaces offers po-tential benefits in terms of efficiency and …

    calgary Repository record for Learning seismic wave propagation with Fourier Neural Operators: Advantages, limitations, and Physics-Informed implementation (opens in a new tab)

  14. Process-Aware Hybrid Deep Learning for Environmental Systems and Extremes

    … suffer from poor predictive performance. Deep learning is a relatively new paradigm that excels at capturing complex patterns in environmental datasets but lacks physical constraints, making it prone to overfitting and poor generalisation. This thesis bridges these paradigms by developing a …

    unsw Repository record for Process-Aware Hybrid Deep Learning for Environmental Systems and Extremes (opens in a new tab)

  15. Data-driven approach and analysis for electronic structure in moiré materials.

    … local environment of each orbital pair. Part II--Operator Learning for Moiré Materials In Part II of the thesis, to identify new moiré materials that may exhibit superconductivity, we have developed a data-driven computational framework that utilizes AI techniques to predict the local density of …

    umn Repository record for Data-driven approach and analysis for electronic structure in moiré materials. (opens in a new tab)

  16. Reduced-set models for improving the training and execution speed of kernel methods

    … of kernel methods, which are a class of machine learning methods known for their wide applicability and state-of-the-art performance, but which suffer from high training and evaluation complexity. The work in this thesis utilizes the notion of reduced-set models to alleviate the training and …

    gatech Repository record for Reduced-set models for improving the training and execution speed of kernel methods (opens in a new tab)

  17. Physics-Informed Machine Learning for Electromagnetic and Multiphysics Modeling

    … stability requirements. Scientific machine learning (SciML) offers a promising path to accelerate such simulations. However, conventional supervised approaches require large labeled datasets and often generalize poorly outside the training domain. Physics-informed machine learning (PIML) …

    toronto-retro Repository record for Physics-Informed Machine Learning for Electromagnetic and Multiphysics Modeling (opens in a new tab)

  18. Deep Learning For Surrogate Modeling And Uncertainty Quantification In Science & Engineering

    Scientific machine learning (SciML) has become an increasingly important tool for constructing surrogate models of complex physical systems, enabling rapid approximation of expensive numerical solvers and supporting tasks such as design optimization, uncertainty analysis, and autonomous …

    penn Repository record for Deep Learning For Surrogate Modeling And Uncertainty Quantification In Science & Engineering (opens in a new tab)

  19. Stochastic Modeling of Parametric and Model-Form Uncertainties in Computational Mechanics: Applications to Multiscale and Multimodel Predictive Frameworks

    … --- which are uncertainties defined at the operator level. Past contributions in the field have primarily focused on parametric uncertainties in idealized environments involving simple state spaces and index sets. On the other hand, the consideration of model-form uncertainties (beyond model …

    duke Repository record for Stochastic Modeling of Parametric and Model-Form Uncertainties in Computational Mechanics: Applications to Multiscale and Multimodel Predictive Frameworks (opens in a new tab)

  20. Learning Macro (and Other) Operators for use Outwith Search to Improve Planner Performance

    … the hypothesis that it is possible to learn operators to add to planning domain models, that are not used within planning search, yet reduce the time taken to solve planning problems. Previous research has shown that macro-operators can help to improve planner performance. However, …

    kings Repository record for Learning Macro (and Other) Operators for use Outwith Search to Improve Planner Performance (opens in a new tab) Download PDF for Learning Macro (and Other) Operators for use Outwith Search to Improve Planner Performance (opens in a new tab)

Page 1 of 2