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 44 for “"OOD Detection"”.

  1. Representation learning for OOD detection and explanation

    … predictions on<br/>out-of-distribution (OOD) data which can lead to severe consequences in safety-critical settings.<br/>Therefore, two fundamental requirements for the deployment of ML models are: 1) being able to<br/>identify OOD data; 2) being able to explain why a data point is …

    bristol Repository record for Representation learning for OOD detection and explanation (opens in a new tab) Download PDF for Representation learning for OOD detection and explanation (opens in a new tab)

  2. Towards Pixel-Level OOD Detection for Semantic Segmentation

    There exists wide research surrounding the detection of out of distribution sample for image classification. Safety critical applications, such as autonomous driving, would benefit from the ability to localise the unusual objects causing an image to be out of distribution. This thesis adapts …

    waterloo Repository record for Towards Pixel-Level OOD Detection for Semantic Segmentation (opens in a new tab)

  3. Improving OOD Detection, Recognition, and Understanding via Multi-Modal Feature Alignment

    … requires robust handling of out-of-distribution (OOD) data—inputs that differ from the training distribution due to semantic shift (e.g., novel object categories) or covariate shift (e.g., changes in lighting or sensor noise). Achieving this robustness requires that AI systems progress from …

    waterloo Repository record for Improving OOD Detection, Recognition, and Understanding via Multi-Modal Feature Alignment (opens in a new tab)

  4. Out-of-Distribution Detection for LiDAR-based 3D Object Detection

    3D object detection is an essential part of automated driving, and deep neural networks (DNNs) have achieved state-of-the-art performance for this task. However, deep models are notorious for assigning high confidence scores to out-of-distribution (OOD) inputs, that is, inputs that are not drawn …

    waterloo Repository record for Out-of-Distribution Detection for LiDAR-based 3D Object Detection (opens in a new tab)

  5. Knowing What We Don't Know! Detecting and Learning Out-of-Distribution Data in the Open World

    … and encountering novel, Out-of-Distribution (OOD) instances that lie beyond the training distribution is the norm. The success of these closed-world AI systems has yet to translate to the open world, where their expertise is lacking, leading to their catastrophic failure. For reliable …

    penn Repository record for Knowing What We Don't Know! Detecting and Learning Out-of-Distribution Data in the Open World (opens in a new tab)

  6. CONFIDENTLY DETECTING OUT-OF-DISTRIBUTION SIGNALS WITH UNCERTAINTY

    … This problem is known as out-of-distribution (OOD) detection. Existing methods commonly rely on deterministic models and statistical machine learning. Not only do deterministic models not recognize and appropriately classify unknown data, but they also do not report their uncertainty in the …

    nps Repository record for CONFIDENTLY DETECTING OUT-OF-DISTRIBUTION SIGNALS WITH UNCERTAINTY (opens in a new tab)

  7. Out-of-distribution detection using inter-level features of deep neural networks

    … addresses the problem of Out-of-Distribution (OOD) detection in deep learning-based classifiers. The emphasis is to investigate real-world settings, where both the In-distribution (ID) and OOD samples are hard to distinguish due to their high semantic similarities. Despite the significant …

    ubc Repository record for Out-of-distribution detection using inter-level features of deep neural networks (opens in a new tab) Download PDF for Out-of-distribution detection using inter-level features of deep neural networks (opens in a new tab)

  8. Exploiting Interclass Relationships for Improved Deep Neural Network Generalization in Open-world Environments

    … This is the subject of out-of-distribution (OOD) detection and open-set recognition research. We introduce two perspectives for solving the OOD detection task. First, we reconsider the OOD detection problem from the hierarchical classification perspective. Second, we examine the problem from …

    duke Repository record for Exploiting Interclass Relationships for Improved Deep Neural Network Generalization in Open-world Environments (opens in a new tab)

  9. KLIP: Localized distribution shift detection via KL-divergence with diffusion priors in inverse problems

    … as some capacity to detect out-of-distribution (OOD) images. However, existing approaches to OOD detection often require some knowledge of the shifted distribution, fail to detect subtle or localized distribution shifts, and operate on full images, rather than the indirect measurements available …

    gatech Repository record for KLIP: Localized distribution shift detection via KL-divergence with diffusion priors in inverse problems (opens in a new tab)

  10. Uncertainty estimation of weakly supervised predictive models for out-of-distribution detection in digital pathology

    … that are regarded as the out-of-distribution (OOD) data, distinct from the data used for model training. Accurate uncertainty estimates can potentially reduce the adverse effects of OOD regions on the target classification task in a clinical workflow. This, in turn, can prevent models from …

    ubc Repository record for Uncertainty estimation of weakly supervised predictive models for out-of-distribution detection in digital pathology (opens in a new tab) Download PDF for Uncertainty estimation of weakly supervised predictive models for out-of-distribution detection in digital pathology (opens in a new tab)

  11. An Application of Out-of-Distribution Detection for Two-Stage Object Detection Networks

    … train time. These so-called out-of-distribution (OOD) detection techniques hold promise for improving safety in systems where unusual or novel inputs may results in errors that endanger human lives. Autonomous vehicles could specifically benefit from the use of these techniques if they could be …

    waterloo Repository record for An Application of Out-of-Distribution Detection for Two-Stage Object Detection Networks (opens in a new tab)

  12. Dideliais kalbų modeliais sugeneruotų vaizdų atpažinimas klasifikavimo uždavinyje /

    … examples are referred as out-of-distribution (OOD). In classification tasks, detecting these distributional shifts is essential to ensure the safety of the models for its users. The study investigates five different out-of-distribution (OOD) detection methods – analyzing their underlying …

    vilnius Repository record for Dideliais kalbų modeliais sugeneruotų vaizdų atpažinimas klasifikavimo uždavinyje / (opens in a new tab)

  13. On Impact of Network Architecture for Deep Learning

    … of CNN architectures on Out-of-Distribution (OOD) detection tasks using Neural Architecture Search (NAS). To improve the quality of evaluation on OOD detection during the search, we propose evolving distillation based on our multi-view feature learning explanation. Experimental results show …

    duke Repository record for On Impact of Network Architecture for Deep Learning (opens in a new tab)

  14. Adaptations of rehearsal methods for online continual learning

    … work integrates two critical mechanisms: drift detection and out-of-distribution (OoD) detection. Drift detection involves identifying shifts in the data distribution—commonly referred to as concept drift—caused by changes in user behavior, sensor degradation, or environmental variation.OoD …

    greece Repository record for Adaptations of rehearsal methods for online continual learning (opens in a new tab)

  15. Uncertainty-aware weather-related road surface condition classification

    … referred to as Out-Of-Distribution (OOD) samples, can occur, which the algorithm then confidently uses to make driving decisions. In a worst-case scenario, this can lead to injuring passengers or property. The primary objective of this thesis is to implement and analyse methods for …

    chalmers Repository record for Uncertainty-aware weather-related road surface condition classification (opens in a new tab)

  16. Investigating Distribution Shifts for Deep Learning Models

    … enhance multiple out-of-distribution (OOD) detection methods by simply incorporating synthetic in-distribution images into the training. Then, we present MixOE, a novel training methodology that significantly improves OOD detection in the challenging (yet practical) fine-grained …

    duke Repository record for Investigating Distribution Shifts for Deep Learning Models (opens in a new tab)

  17. Trustworthy Machine Learning with Deep Generative Models

    … aspects of trustworthiness: out-of-distribution (OOD) detection and privacy preservation. Generative models serve purposes beyond generating realistic samples. Likelihood-based DGMs, e.g. flow generative models, can additionally compute the likelihood of input data, which can be used as an …

    waterloo Repository record for Trustworthy Machine Learning with Deep Generative Models (opens in a new tab)

Page 1 of 3