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 70 for “"Contrastive loss"”.

  1. Contrastive Loss in Recommendation Systems

    … Rank Pairwise (WARP) and Information Noise-Contrastive Estimation (InfoNCE) losses, to optimize the embeddings of users and recommended items in the shape of a ranking list. The results show that GraphSAGE outperforms LightGCN, and more specifically, among GraphSAGE models, the one with the …

    arkansas Repository record for Contrastive Loss in Recommendation Systems (opens in a new tab)

  2. Scheduled Contrastive Loss In Continued Transfer Learning For Software Engineering Tasks

    … software engineering tasks. At the same time, contrastive learning has shown promise in other modalities, such as computer vision-related problems, and has been explored to some extent in terms of limited software engineering tasks. We demonstrate that contrastive loss, on its own, is …

    wm-mary Repository record for Scheduled Contrastive Loss In Continued Transfer Learning For Software Engineering Tasks (opens in a new tab)

  3. Siamese Network with Dynamic Contrastive Loss for Semantic Segmentation of Agricultural Lands

    … Introducing a novel approach termed "dynamic contrastive learning," this research refines the existing contrastive learning framework. Tailored modifications aim to improve the model's accuracy in classifying and segmenting geographic features accurately. Various deep learning models, …

    unt Repository record for Siamese Network with Dynamic Contrastive Loss for Semantic Segmentation of Agricultural Lands (opens in a new tab)

  4. Symmetric implicit geometries under label-imbalances : from cross-entropy to supervised contrastive loss

    … fit the training data), traditional methods like loss reweighting or data resampling become less effective. While new methods address imbalanced learning in the over-parameterized regime, their impact on deep learning is unclear. Recent studies reveal the Neural Collapse (NC) phenomenon in …

    ubc Repository record for Symmetric implicit geometries under label-imbalances : from cross-entropy to supervised contrastive loss (opens in a new tab) Download PDF for Symmetric implicit geometries under label-imbalances : from cross-entropy to supervised contrastive loss (opens in a new tab)

  5. Exploring Loss Functions in Machine Learning

    The loss function plays a critical role in machine learning. It is fundamental in training, evaluating, and optimizing machine learning models, directly impacting their effectiveness and efficiency in solving specific tasks. We explore three new loss functions and their applications. Softmax …

    claremont Repository record for Exploring Loss Functions in Machine Learning (opens in a new tab)

  6. Contrastive Learning, with Application to Forensic Identification of Source

    … advances in deep learning have been made by contrastive learning methods. Many forensic identification of source problems deal with a scarcity of data, an issue addressed by few-shot learning. In this work, we make specific what makes a neural network a contrastive network. We then consider …

    sdstate Repository record for Contrastive Learning, with Application to Forensic Identification of Source (opens in a new tab)

  7. Aplikace pro automatické vyhodnocení věrohodnosti generovaného snímku obličeje

    … sítích, které používají ztrátovou funkci contrastive loss. Model byl trénován a testován na datové sadě LFW, kde dosáhl přesnosti až 91 %. Pro testování generovaných snímků je použit model StarGAN, který generoval snímky obličeje se změnou barvy vlasů, pohlaví a stáří. Výsledné testování …

    brno-tech Repository record for Aplikace pro automatické vyhodnocení věrohodnosti generovaného snímku obličeje (opens in a new tab)

  8. Aplikace pro automatické vyhodnocení věrohodnosti generovaného snímku obličeje

    … sítí, u kterých používá ztrátovou funkci contrastive loss. Tento model byl trénován a testován na datové sadě LFW, kde se dostal až k přesnosti 91 %. Pro testování generovaných snímků je použit model StarGAN, který generoval snímky obličeje se změnou barvy vlasů, pohlaví a stáří. Výsledné …

    brno-tech Repository record for Aplikace pro automatické vyhodnocení věrohodnosti generovaného snímku obličeje (opens in a new tab)

  9. Self-Supervised Remote Sensing Image Change Detection and Data Fusion

    … proposes a bi-temporal / multi-temporal contrastive change detection framework, which employs contrastive loss on image patches or superpixels to get fine-grained change maps and incorporates an uncertainty method to enhance the temporal robustness. In the context of satellite image time …

    trento Repository record for Self-Supervised Remote Sensing Image Change Detection and Data Fusion (opens in a new tab)

  10. Predictive Models for Multi-Modality Electronic Healthcare Records

    … stamps during an ICU stay. (2) A multi-modal contrastive learning framework for diagnosis prediction (MCDP), which uses the multi-modal contrastive loss to boost the prediction performance in healthcare. The framework uses the diagnosis codes and clinical features as two related modalities. We …

    buffalo Repository record for Predictive Models for Multi-Modality Electronic Healthcare Records (opens in a new tab)

  11. PERSONALIZED VISUAL INFORMATION CAPTIONING

    … graphs and a model leveraging soft label contrastive loss for effective feature extraction. Experimental results validate the competitiveness of the proposed methods across various datasets.

    nus Repository record for PERSONALIZED VISUAL INFORMATION CAPTIONING (opens in a new tab) Download file for PERSONALIZED VISUAL INFORMATION CAPTIONING (opens in a new tab)

  12. Towards Multi-modal Explainable Video Understanding

    … of this work is the introduction of a novel VL contrastive loss function. This function plays an essential role in ensuring that the learned embedding features are semantically consistent with the video captions. By aligning the embeddings with the ground truth captions, the VL contrastive loss …

    arkansas Repository record for Towards Multi-modal Explainable Video Understanding (opens in a new tab)

  13. Vision transformer fine-tuning for image similarity using metric learning /

    … investigates how different metric learning loss functions affect the fine­tuning of a DINOv2 Vision Transformer for image similarity tasks. Experiments are conducted on the DISC21 dataset, a large­scale benchmark for image copy detection. Five loss functions—contrastive, triplet, ArcFace, …

    vilnius Repository record for Vision transformer fine-tuning for image similarity using metric learning / (opens in a new tab)

  14. Yet another image clustering framework using deep learning

    … from deep embeddings by combining instance-level contrastive learning with a deep embedding based cluster center predictor. Our approach jointly learns representations and predicts cluster centers in an end-to-end manner. This is accomplished via a three-pronged approach that combines a clustering …

    iastate Repository record for Yet another image clustering framework using deep learning (opens in a new tab)

  15. Self-supervised Representation Learning in Computer Vision and Reinforcement Learning

    … representation learning (SSL). We consider both contrastive and non-contrastive methods and present a new loss function for SSL based on feature whitening. Our solution is conceptually simple and competitive with other methods. Self-supervised representations are beneficial for most areas of deep …

    trento Repository record for Self-supervised Representation Learning in Computer Vision and Reinforcement Learning (opens in a new tab)

  16. Enhancing contrastive representation learning through data

    Contrastive learning learns input representation by pushing similar data together and pulling dissimilar data away, along with data augmentation and pretext task construction. It enhances the large model learning due to its ability to use a large amount of unlabeled data. It has been successfully …

    hawaii Repository record for Enhancing contrastive representation learning through data (opens in a new tab)

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

    … extraction method. We also propose to use a contrastive loss to improve both the performance of the object detection and OOD detection methods. To evaluate OOD detection methods, we develop a simple but effective technique of generating OOD objects for a given object detection model. Our …

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

  18. Vision transformer-based context-aware system for lingual ultrasound in digital health ecosystem

    … encoder-decoder convolutional neural network, contrastive loss and quality control process supported with human-reinforcement feedback to extract tongue fingerprint. Second, design ZTCloudGuard for access control within the telehealth cloud-based eco-system between. The architecture manage …

    uvic Repository record for Vision transformer-based context-aware system for lingual ultrasound in digital health ecosystem (opens in a new tab)

  19. Siamese Network-Based Multi-Modal Deepfake Detection

    … deepfake model with invariant of constructive loss and triplet loss. Contrastive loss uses the trained network's output for a positive example and calculates its distance to an instance of the same class and contrasts it with the range to negative samples. The triplet loss was computed by …

    umkc Repository record for Siamese Network-Based Multi-Modal Deepfake Detection (opens in a new tab)

  20. Human-Centric Deep Generative Models: The Blessing and The Curse

    … I probe on applying attention modules and dual contrastive loss to generative adversarial networks (GANs), which pushes photorealistic image generation to a new state of the art. For the steerability, I introduce Texture Mixer, a simple yet effective approach to achieve steerable texture …

    maryland Repository record for Human-Centric Deep Generative Models: The Blessing and The Curse (opens in a new tab)

Page 1 of 4