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 46 for “"Pseudo Labeling"”.

  1. Federated Medical Image Segmentation with Pseudo-Labeling under Model Heterogeneity

    … public dataset. The server applies adaptive pseudo-label aggregation to obtain high-quality pseudo-labels. A two-stage filtering mechanism— i.e., uncertainty-based filtering followed by SSL-guided similarity filtering—selects personalized pseudo-labeled subsets aligned with each client’s data …

    calgary Repository record for Federated Medical Image Segmentation with Pseudo-Labeling under Model Heterogeneity (opens in a new tab)

  2. Rice Panicle Segmentation using Multi-Stage Pseudo-labeling and Active Learning

    … conditions, and the high cost of manual labeling. This research aims to alleviate these challenges by proposing accurate and label-efficient deep learning frameworks tailored for rice panicle segmentation under real-world agricultural settings. A novel dataset was developed using …

    calgary Repository record for Rice Panicle Segmentation using Multi-Stage Pseudo-labeling and Active Learning (opens in a new tab)

  3. Transformer-Based Multi-Object Tracking of Football Players Using Pseudo-Labeling

    … football-specific tracking data, we implement a pseudo-labeling framework that allows the model to be trained on unlabelled domain data in a semi-supervised fashion. This approach enables progressive refinement of the model through multiple training cycles on domain-specific content. Our results …

    chalmers Repository record for Transformer-Based Multi-Object Tracking of Football Players Using Pseudo-Labeling (opens in a new tab)

  4. Federated Pseudo-labeling: A Data-Centric, Privacy-Preserving Framework for Medical Image Segmentation

    … sharing entirely. DCFed leverages federated pseudo-labeling on publicly available unlabeled datasets, using logit-based aggregation and uncer-tainty estimation. Clients independently train on their private datasets while collaboratively generating and refining pseudo-labels on a shared public …

    calgary Repository record for Federated Pseudo-labeling: A Data-Centric, Privacy-Preserving Framework for Medical Image Segmentation (opens in a new tab)

  5. Novel Algorithms for Interpretable and Semi-Supervised Machine Learning

    Pseudo-labeling (PL) has historically been considered synonymous with model-provided weak supervision in the context of semi-supervised learning (SSL). In that context, pseudo-labels are simply an inferior version of human supervision. Framing pseudo-labeling this way provides an overly restrictive …

    oklahoma-thes Repository record for Novel Algorithms for Interpretable and Semi-Supervised Machine Learning (opens in a new tab)

  6. Reducing the Cost of Test Data Labelling for Deep-Learning Systems: An Empirical Study

    … study aimed at reducing the cost of test data labeling for deep-learning systems through automated and semi-automated methods. We propose two approaches: Human-in-the-Loop labeling and Pseudo Labeling. The Human-in-the-Loop approach combines machine learning with human expertise, intervening …

    ottawa-retro Repository record for Reducing the Cost of Test Data Labelling for Deep-Learning Systems: An Empirical Study (opens in a new tab)

  7. Semi-supervised učení pro rozpoznávání textu

    … manuskript. Zvolenou semi-supervised metodou je pseudo-labeling v kombinaci s filtrováním na základě konfidence modelu, která byla aplikována na neuronovou síť architektury Transformer. Experimentálně byly ověřeny podmínky, kdy je tato metoda účinná, a bylo dosaženo pozitivních výsledků. Přínosem …

    brno-tech Repository record for Semi-supervised učení pro rozpoznávání textu (opens in a new tab)

  8. Remote multimodal sensor monitoring and analytics for aging and neurodegenerative clinical assessment support

    … disease populations. However, the labeling of high-frequency sensor observations with sparsely collected functional assessment instruments requires developing integrative methodological approaches for supervised learning to support clinical decision-making. This dissertation …

    missouri Repository record for Remote multimodal sensor monitoring and analytics for aging and neurodegenerative clinical assessment support (opens in a new tab) Embargoed until 2026-12-01

  9. Annotation Efficient Visual Recognition: from Semi-Supervised to Few-Shot Learning

    … in supervised learning, we introduce a pseudo-labeling framework for semi-supervised learning. While consistency regularization methods dominate the field, they heavily rely on domain-specific data augmentations, limiting their applicability. We argue that even though pseudo-labeling is …

    ucf Repository record for Annotation Efficient Visual Recognition: from Semi-Supervised to Few-Shot Learning (opens in a new tab)

  10. SEMI-SUPERVISED AND ADVERSARIAL DATA SYNTHESIS FOR LANGUAGE MODELING

    … few-shot learning framework integrating pseudo-labeling and self-supervised objectives to reduce fine-tuning data needs; (2) a semi-supervised strategy for sentence representation learning, improving synthetic data quality for entailment and contradiction tasks; (3) adversarial data …

    nus Repository record for SEMI-SUPERVISED AND ADVERSARIAL DATA SYNTHESIS FOR LANGUAGE MODELING (opens in a new tab) Download file for SEMI-SUPERVISED AND ADVERSARIAL DATA SYNTHESIS FOR LANGUAGE MODELING (opens in a new tab)

  11. Semi-Supervised Galaxy Morphology Classification Using Convolutional Vision Transformers

    … to the new imaging domain and refined using pseudo-labeling on unlabeled data. Experiments simulating JADES imaging conditions show that reduced resolution and noise significantly impact the separability of certain morphology classes, particularly edge-on and cigar-shaped smooth galaxies. …

    south-carolina Repository record for Semi-Supervised Galaxy Morphology Classification Using Convolutional Vision Transformers (opens in a new tab)

  12. Mitigating the challenges of distribution shift under strong computational constraints

    … alignment, feature-level alignment, and pseudo-labeling is proposed for cross-domain object detection, highlighting the power of unsupervised adaptation techniques. The thesis also extends its investigation to online continual learning with label delay, proposing a novel approach to …

    oxford Repository record for Mitigating the challenges of distribution shift under strong computational constraints (opens in a new tab)

  13. Assessing the feasibility of applying machine learning tools to predict environmental stressors from digital urban fingerprints

    … (GNN) framework with domain adaptation and pseudo-labeling, enabling robust noise mapping across five UK cities using standardized remote sensing and Urban Atlas LULC data.<br/>Results demonstrate that deep learning models—when properly integrated with remote sensing and urban structural …

    soton Download PDF for Assessing the feasibility of applying machine learning tools to predict environmental stressors from digital urban fingerprints (opens in a new tab)

  14. Avaliação do custo-benefício da aplicação de métodos de aprendizado semi supervisionado para detecção de Equipamentos de Proteção Individual em Ambiente Industrial

    … 10%, 20% e 30% dos dados, aplicando técnicas de pseudo-labeling através da arquitetura YOLOv8. Os resultados foram avaliados mediante validação cruzada com k=10 repetições e análise estatística usando testes de Friedman e Nemenyi. Os modelos semi supervisionados demonstraram performance …

    brazil-ufpe Repository record for Avaliação do custo-benefício da aplicação de métodos de aprendizado semi supervisionado para detecção de Equipamentos de Proteção Individual em Ambiente Industrial (opens in a new tab)

  15. Towards Label Efficiency and Privacy Preservation in Video Understanding

    … actions. To address this, we propose "FinePseudo" framework, which leverages temporal alignability to learn phase-aware distances. It also introduces collaborative pseudo-labeling between video-level and alignability encoder, refining the pseudo-labeling process for fine-grained actions. …

    ucf Repository record for Towards Label Efficiency and Privacy Preservation in Video Understanding (opens in a new tab)

  16. Advancing Differential Equation-based Modeling in Computational Medicine and Deep Knowledge Transfer for Multi-Modal GeoAI

    … masked autoencoding and soft-alignment pseudo-labeling to fine-tune foundational models such as a recent 2023 NASA/IBM geospatial foundation model. This approach improves segmentation accuracy and adaptability across RS modalities, including RGB, multispectral, and hyperspectral imagery. …

    houston Repository record for Advancing Differential Equation-based Modeling in Computational Medicine and Deep Knowledge Transfer for Multi-Modal GeoAI (opens in a new tab) Embargoed until 2027-05-01

  17. Driver Behavior Classification in Electric Vehicles

    … query by committee, active deep dropout, gradual pseudo labeling, and active learning for time-series data. These techniques showed variable results, with uncertainty sampling performing consistently better than random sampling. This study confirms the effectiveness of machine learning models in …

    chalmers Repository record for Driver Behavior Classification in Electric Vehicles (opens in a new tab)

  18. Data Quality Evaluation and Improvement for Machine Learning

    … for data quality improvement. Compared to pseudo-labeling, co-training, and expectation-maximization (EM), generative adversarial network (GAN) is more effective for automated data augmentation, especially when a small portion of labeled data and a large amount of unlabeled data is …

    unt Repository record for Data Quality Evaluation and Improvement for Machine Learning (opens in a new tab)

  19. Exploring semi-supervised methods in magnetic resonance imaging analysis /

    … methods, namely consistency regularization, pseudo-labeling, co-training, contrastive learning, adversarial learning, and hybrid methods. Comparative experiments were conducted, with techniques including Mean Teacher (MT), Deep Adversarial Networks (DAN), Adversarial Entropy Minimization …

    vilnius Repository record for Exploring semi-supervised methods in magnetic resonance imaging analysis / (opens in a new tab)

Page 1 of 3