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 38 for “"Generalization and Robustness"”.

  1. FROM PARSIMONY TO LOCAL SENSITIVITY: GENERALIZATION AND ROBUSTNESS IN MODERN MACHINE LEARNING

    … settings. As a result, a rigorous understanding of their capabilities and limitations is essential. This thesis investigates two complementary paradigms of evaluation in supervised classification: generalization and robustness. Generalization captures a model’s ability to extrapolate to …

    jhu Repository record for FROM PARSIMONY TO LOCAL SENSITIVITY: GENERALIZATION AND ROBUSTNESS IN MODERN MACHINE LEARNING (opens in a new tab)

  2. ON THE GENERALIZATION AND ROBUSTNESS OF ML-BASED NETWORK INTRUSION DETECTION SYSTEMS

    … application, concerns persist regarding the generalization and robustness of ML-based NIDS. To facilitate real-world deployment, it is crucial to ensure the generalization and robustness of ML-based NIDS to attain a certain level of effectiveness. The current practices for ML-based NIDS are …

    siu-theses Repository record for ON THE GENERALIZATION AND ROBUSTNESS OF ML-BASED NETWORK INTRUSION DETECTION SYSTEMS (opens in a new tab)

  3. Building digital twins of articulated objects and scenes through interactive perception

    Perceiving and interacting with complex household environments with human-level robustness and flexibility is an open challenge in embodied AI and robotics. Despite significant progress in AI-powered robotics, manipulation algorithms currently lack the necessary generalization and robustness …

    texas Repository record for Building digital twins of articulated objects and scenes through interactive perception (opens in a new tab)

  4. Combine and conquer: representation learning from multmodal data

    … is becoming less popular due to its high cost and various issues with generalization and robustness. This is unsurprising, as data such as images and language are complex and cannot be accurately represented by a single label. When models are trained using this method, they often learn features …

    oxford Repository record for Combine and conquer: representation learning from multmodal data (opens in a new tab)

  5. Kernel Perspectives on Generalization in High Dimensions

    … how feature representations influence generalization and robustness in high-dimensional and overparameterized machine learning (ML) models. Using classical linear and kernel methods as a starting point, we first demonstrate that these simpler models can exhibit surprising optimization …

    gatech Repository record for Kernel Perspectives on Generalization in High Dimensions (opens in a new tab)

  6. Final energy consumption in road transport by means of transport /

    … supplementary data, harmonizes diverse sources, and uses predictive machine learning models. The study defines the steps of data cleaning, preprocessing, modeling, and hyperparameter tuning while proposing a unified framework adapted to different vehicle types, such as passenger cars and vans. …

    vilnius Repository record for Final energy consumption in road transport by means of transport / (opens in a new tab)

  7. Optimizing Deep Neural Networks: Leveraging Structural Characteristics for Enhanced NAS and Adversarial Robustness

    … in computer vision, natural language processing, and autonomous systems. However, the predominant focus on training techniques, data scaling, and computational optimizations often neglects the structural design of models, which is fundamental to both generalized and adversarial accuracy, …

    duke Repository record for Optimizing Deep Neural Networks: Leveraging Structural Characteristics for Enhanced NAS and Adversarial Robustness (opens in a new tab)

  8. Improving classifier robustness

    … are the state-of-the-art in classification and increasingly deployed in domains where safety and security are critical. Such domains include banking, autonomous driving, malware detection, and cancer detection. Unlike humans, machine learning classifiers can be easily fooled by attacks that …

    aalto Repository record for Improving classifier robustness (opens in a new tab)

  9. Robustness and generalization guarantees for statistical learning of generative models

    … the context of generative models. By combining standard methods based on the theory of empirical processes with ideas from optimal transport and signal recovery, we formally address the generalization and robustness guarantees for the existing and newly suggested algorithms. More specifically, we …

    uiuc Repository record for Robustness and generalization guarantees for statistical learning of generative models (opens in a new tab)

  10. Fadiga de alto ciclo em compósitos de PRFV. Modelagem por RNAs e prevenção de falha

    … takes the shape of a short fiber mats and a bidirectional textile fabric which originates composite laminates ranging from 10 to 12 layers. The laminates were submitted to tension, compression and high cycle fatigue efforts and all these procedure were carried out by considering a …

    brazil-ufrn Repository record for Fadiga de alto ciclo em compósitos de PRFV. Modelagem por RNAs e prevenção de falha (opens in a new tab)

  11. On Modeling Long-Range Dependencies for Visual Perception

    … from visual inputs. An example is to recognize and segment objects from natural images. Recently, deep networks enable us to do a wide range of these tasks better than ever. These are mostly achieved with convolutional neural networks that model pixel relations within a small convolution kernel. …

    jhu Repository record for On Modeling Long-Range Dependencies for Visual Perception (opens in a new tab)

  12. Reinforcement learning in large state action spaces

    … task constraints like decentralization and lack of guarantees about important properties like performance, generalization and robustness in potentially unseen scenarios.</p> <p>This thesis is motivated towards bridging the aforementioned gap. We propose several principled algorithms and …

    oxford Repository record for Reinforcement learning in large state action spaces (opens in a new tab)

  13. Understanding Feature Learning and Calibration of Machine Learning Models Through Simple Data Distributions

    … to perform well requires a variety of tricks and heuristics, most of which are introduced with a non-rigorous explanation of why they work.</p><p>In the first part of this thesis, we peer more deeply into when and why a particular class of such heuristics - namely data augmentation - can …

    duke Repository record for Understanding Feature Learning and Calibration of Machine Learning Models Through Simple Data Distributions (opens in a new tab)

  14. A Transformer-Based Perception and Analytics Framework for Cattle Behavior Analysis across Diverse Ranch Environments

    … important information about animal health and welfare. Visionbased monitoring systems provide a non-contact means of observing these behaviors at scale. However, most existing systems are designed for controlled settings and perform poorly when applied to new ranch environments. Variations …

    texas-state Repository record for A Transformer-Based Perception and Analytics Framework for Cattle Behavior Analysis across Diverse Ranch Environments (opens in a new tab)

  15. Out-of-domain few-shot learning for image classification

    … trained on significant amounts of labeled data, generalization notably suffers when using relatively small datasets or testing on out-of-domain classes. Acquiring new, labelled samples, however, can be an expensive or impractical affair in some situations. In an attempt to address the issue of …

    ubc Repository record for Out-of-domain few-shot learning for image classification (opens in a new tab) Download PDF for Out-of-domain few-shot learning for image classification (opens in a new tab)

  16. Towards Trustworthy Deep Learning and Large Language Models

    … high-stakes domains such as science, healthcare, and social decision-making, ensuring their trustworthiness becomes critical. Beyond predictive accuracy, trustworthy deep learning requires fairness, robustness, generalization, and interoperability, particularly for geometric deep learning models …

    tamu Repository record for Towards Trustworthy Deep Learning and Large Language Models (opens in a new tab)

  17. Towards Trustworthy Deep Learning and Large Language Models

    … high-stakes domains such as science, healthcare, and social decision-making, ensuring their trustworthiness becomes critical. Beyond predictive accuracy, trustworthy deep learning requires fairness, robustness, generalization, and interoperability, particularly for geometric deep learning models …

    tamu Repository record for Towards Trustworthy Deep Learning and Large Language Models (opens in a new tab)

  18. Towards Trustworthy Deep Learning and Large Language Models

    … high-stakes domains such as science, healthcare, and social decision-making, ensuring their trustworthiness becomes critical. Beyond predictive accuracy, trustworthy deep learning requires fairness, robustness, generalization, and interoperability, particularly for geometric deep learning models …

    tamu Repository record for Towards Trustworthy Deep Learning and Large Language Models (opens in a new tab)

  19. Towards Trustworthy Deep Learning and Large Language Models

    … high-stakes domains such as science, healthcare, and social decision-making, ensuring their trustworthiness becomes critical. Beyond predictive accuracy, trustworthy deep learning requires fairness, robustness, generalization, and interoperability, particularly for geometric deep learning models …

    tamu Repository record for Towards Trustworthy Deep Learning and Large Language Models (opens in a new tab)

  20. Towards Trustworthy Deep Learning and Large Language Models

    … high-stakes domains such as science, healthcare, and social decision-making, ensuring their trustworthiness becomes critical. Beyond predictive accuracy, trustworthy deep learning requires fairness, robustness, generalization, and interoperability, particularly for geometric deep learning models …

    tamu Repository record for Towards Trustworthy Deep Learning and Large Language Models (opens in a new tab)

Page 1 of 2