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 22 for “"Model Inversion"”.

  1. Extreme imaging via physical model inversion : seeing around corners and imaging black holes

    Imaging often plays a critical role in advancing fundamental science. However, as science continues to push the boundaries of knowledge, imaging systems are reaching the limits of what can be measured using traditional-direct approaches. By designing systems that tightly integrate novel sensor and …

    mit Repository record for Extreme imaging via physical model inversion : seeing around corners and imaging black holes (opens in a new tab)

  2. Evolving Threats and Defenses in Machine Learning: Focus on Model Inversion and Beyond

    Machine learning (ML) models are increasingly integrated into critical real-world applications, raising concerns about security, privacy, and trustworthiness. Among various emerging threats, model inversion (MI) attacks stand out due to their potential to compromise the confidentiality of training …

    vt Repository record for Evolving Threats and Defenses in Machine Learning: Focus on Model Inversion and Beyond (opens in a new tab)

  3. Remote sensing of montane forest structure and biomass : a canopy relectance model inversion approach

    lethbridge

  4. Nonlinear Mr Model Inversion for Semi-Active Control Enhancement With Open-Loop Force Compensation

    … Laboratory testing of the hysteretic inversion process was performed with the goal of emulating an ideal linear damper without hysteresis. These results are compared with the implicit assumption thus providing a basis for validating the benefits of the improved methodology.

    vt Repository record for Nonlinear Mr Model Inversion for Semi-Active Control Enhancement With Open-Loop Force Compensation (opens in a new tab)

  5. Data Centric Defenses for Privacy Attacks

    … sensitive information about the data used in model training. These attacks called privacy attacks, exploit the model training process. Contemporary defense techniques make alterations to the training algorithm. Such defenses are computationally expensive, cause a noticeable privacy-utility …

    vt Repository record for Data Centric Defenses for Privacy Attacks (opens in a new tab)

  6. Exploring Hyperspectral and Very High Spatial Resolution Imagery in Vegetation Characterization

    … 1) integration of contribution theory into a model inversion approach to obtain high accuracy in canopy biophysical parameter estimation; 2) exploration and adoption of tree crown longitudinal profiles to achieve high accuracy in tree species classification; and 3) evaluation of canopy health …

    york Repository record for Exploring Hyperspectral and Very High Spatial Resolution Imagery in Vegetation Characterization (opens in a new tab)

  7. Inverse modelling and inverse simulation for system engineering and control applications

    … qualities and associated issues in terms of model validation. However, the available methods still have some well-known limitations. The traditional methods based on the Newton-Raphson algorithm suffer from numerical problems such as high-frequency oscillations and can have limitations in …

    glasgow Repository record for Inverse modelling and inverse simulation for system engineering and control applications (opens in a new tab)

  8. Shifting Paradigms: Data-Centric Approach for Marine Statics Correction using Symmetric Autoencoding

    … correcting marine statics has been based on a model-centric paradigm. This paradigm involves a series of transformations between non-commensurate spaces: first, inversion from seismic data space to velocity model space and second, forward modeling from velocity model space to seismic data …

    mit Repository record for Shifting Paradigms: Data-Centric Approach for Marine Statics Correction using Symmetric Autoencoding (opens in a new tab)

  9. Building Trustworthy Artificial Intelligence of Things Systems in Adversarial Environments

    … adversarial attacks on multimodal diffusion models, motivated by the growing popularity of generative AI technologies. In Chapter 4, we introduce our customized model inversion attack against the medical FL systems. Our attack can reconstruct sensitive real-life COVID-19 X-ray images, brain …

    vt Repository record for Building Trustworthy Artificial Intelligence of Things Systems in Adversarial Environments (opens in a new tab)

  10. Structured learning and inference with neural networks and generative models

    Neural networks and probabilistic models have different and in many ways complementary strengths and weaknesses: neural networks are flexible and support efficient inference, but rely on large quantities of labeled training data. Probabilistic models can learn from fewer examples, but in many cases …

    mit Repository record for Structured learning and inference with neural networks and generative models (opens in a new tab)

  11. Unveiling the impact of neuromotor disorders on speech: a structured approach combining biomechanical fundamentals and statistical machine learning

    … of NDs. This work aims to introduce a working model that is capable of linking both domains and serves as a projection tool to provide insights about a speaker’s neuromotor state. This is based on a review of the neurophysiological background of the structure and function of the nervous system, …

    edinburgh Repository record for Unveiling the impact of neuromotor disorders on speech: a structured approach combining biomechanical fundamentals and statistical machine learning (opens in a new tab)

  12. Privacy-Preserving Natural Language Dataset Generation

    … which attacks such as membership inference or model inversion can extract potentially sensitive training data given the model alone. To prevent curious or malevolent users from gleaning training data through these attacks, we propose the generation of private synthetic datasets to replace the …

    mit Repository record for Privacy-Preserving Natural Language Dataset Generation (opens in a new tab)

  13. Application of maximal information coefficient and affinity propagation to characterizing seismic time series associated with earthquakes

    … feature-based datasets from seismic signals for model inversion and improved subsurface characterization.

    mit Repository record for Application of maximal information coefficient and affinity propagation to characterizing seismic time series associated with earthquakes (opens in a new tab)

  14. Privacy Attacks and Defenses under Security Threats in Machine Learning

    … performance. However, machine learning models exhibit some vulnerabilities against threats in the real world, including privacy risks and security concerns. In terms of privacy risks, malicious users can steal the private information of other users or model owners, including recovering …

    uts Repository record for Privacy Attacks and Defenses under Security Threats in Machine Learning (opens in a new tab)

  15. Direct Waveform Inversion Using Explicit Time-Space Causality Principle

    I have developed a Direct Waveform Inversion (DWI) scheme to simultaneously address several existing challenges in full waveform inversion (FWI). A key ingredient in the DWI is the explicit use of the wavefield time-space causality property in the inversion, which allows us to convert the global …

    houston Repository record for Direct Waveform Inversion Using Explicit Time-Space Causality Principle (opens in a new tab)

  16. Toward sequence-to-structure predictions of chromatin: Generative AI sheds light on genome organization

    … methodological advances. An efficient Hi-C inversion algorithm appears first. This technique extracts pairwise contact potentials from experimental Hi-C data, uncovering mechanistic details obscured by the correlation between Hi-C contact probabilities. This required the development of a …

    mit Repository record for Toward sequence-to-structure predictions of chromatin: Generative AI sheds light on genome organization (opens in a new tab)

  17. Multi-variable optimal numerical control using adaptive model for identification of thermally induced deformation in high-speed machine tools

    … the use of true feedback control. Process models relating the thermal deformation to the temperature rise at some points on the structure are frequently used, but since complicated models are not practical in a real-time control environment, simplified empirical models of the structure are …

    concordia Repository record for Multi-variable optimal numerical control using adaptive model for identification of thermally induced deformation in high-speed machine tools (opens in a new tab)

  18. Three-dimensional Integral Boundary Layer Method for Viscous Aerodynamic Analysis

    … contributions in both the physical and numerical modeling aspects. First, this thesis presents novel closure modeling strategies for 3D IBL and develops a new set of closure models, which were lacking in previous 3D IBL methods. Original 3D boundary layer data sets have been generated and form the …

    mit Repository record for Three-dimensional Integral Boundary Layer Method for Viscous Aerodynamic Analysis (opens in a new tab)

  19. Crustal deformation source monitoring using advanced InSAR time series and time dependent inverse modeling

    … I present novel static and time dependent model inversion approaches. Almost any interferograms include areas where the signal decorrelates and is distorted by atmospheric delay. In this thesis I detail new analysis methods to reduce the limitations of conventional InSAR, by combining the …

    potsdam-diss Repository record for Crustal deformation source monitoring using advanced InSAR time series and time dependent inverse modeling (opens in a new tab)

  20. Adversarial Resilient and Privacy Preserving Deep learning

    … intelligence, ranging from data poisoning and model inversion during the training phase and adversarial evasion attacks during model inference phase, aiming to cause the well-trained model to misbehave randomly or purposefully. This dissertation research addresses these problems with dual …

    gatech Repository record for Adversarial Resilient and Privacy Preserving Deep learning (opens in a new tab)

Page 1 of 2