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 73 for “"Contrastive Learning"”.

  1. Contrastive Learning of Auditory Representations

    Learning rich visual representations using contrastive self-supervised learning has been extremely successful. However, it is still a major question whether we could use a similar approach to learn more efficient auditory and audio-visual representations. In this thesis, we expand on prior …

    uwo Repository record for Contrastive Learning of Auditory Representations (opens in a new tab)

  2. Adversarial graph contrastive learning with information regularization

    Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-11-11 without embargo terms

    uiuc Repository record for Adversarial graph contrastive learning with information regularization (opens in a new tab)

  3. Dynamics of Gradient Flow with Contrastive Learning

    Contrastive learning (CL), in di erent forms, has been shown to learn discriminatory representations for downstream tasks without the need of human labeling. In the representation space learnt via CL, each class collapses to a distinct vertex of a simplex on a hypersphere during training. This …

    mit Repository record for Dynamics of Gradient Flow with Contrastive Learning (opens in a new tab)

  4. Improving Natural Language Understanding via Contrastive Learning Methods

    … Among the previous NLU solutions, representation learning methods have recently become the mainstream, which maps textual data into low-dimensional vector spaces for downstream tasks. With the development of deep neural networks, text representation learning has achieved state-of-the-art …

    duke Repository record for Improving Natural Language Understanding via Contrastive Learning Methods (opens in a new tab)

  5. Towards General-purpose Vision via Multiview Contrastive Learning

    Representation learning plays a key role in building robust and general-purpose vision learners, and is a long-standing problem. It becomes increasingly interesting with the continuing explosion of data in our era. However, most previous approaches are based on specific designs of strategies that …

    mit Repository record for Towards General-purpose Vision via Multiview Contrastive Learning (opens in a new tab)

  6. Uncertainty-Inclusive Contrastive Learning for Leveraging Synthetic Images

    … synthesized training data to improve few-shot learning performance. Prevailing approaches treat all generated data as uniformly important, neglecting the fact that the quality of generated images varies across different domains, datasets, and methods of generation. Using poor-quality images can …

    mit Repository record for Uncertainty-Inclusive Contrastive Learning for Leveraging Synthetic Images (opens in a new tab)

  7. Scaling contrastive learning batch size by two orders of magnitude

    Contrastive learning has emerged as a powerful framework for unsupervised representation learning, allowing models to learn by maximizing agreement between related samples and distinguishing dissimilar ones. However, contrastive learning frameworks are fundamentally limited by the number of …

    mit Repository record for Scaling contrastive learning batch size by two orders of magnitude (opens in a new tab)

  8. Revisiting Contrastive Learning through the Lens of Neighborhood Component Analysis

    … a seminal tool in self-supervised representation learning, contrastive learning has gained unprecedented attention in recent years. In essence, contrastive learning aims to leverage pairs of positive and negative samples for representation learning, which relates to exploiting neighborhood …

    mit Repository record for Revisiting Contrastive Learning through the Lens of Neighborhood Component Analysis (opens in a new tab)

  9. LLM2IR: simple unsupervised contrastive learning makes long-context LLM great retriever

    Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2025-10-19 without embargo terms

    uiuc Repository record for LLM2IR: simple unsupervised contrastive learning makes long-context LLM great retriever (opens in a new tab)

  10. Enhancing Breast Cancer Detection Through Combination of Contrastive Learning and Adversarial Domain Adaptation

    … and therapies. This study aims to explore deep-learning techniques that can be utilized to implement and train a model to identify breast cancer cases in mammograms. Current deep learning-based diagnostic techniques are hindered by two fundamental issues: the expensive and time-consuming task of …

    windsor Repository record for Enhancing Breast Cancer Detection Through Combination of Contrastive Learning and Adversarial Domain Adaptation (opens in a new tab)

  11. Graph Neural Networks for Health-Aware Food and Multi-Criteria Recommendation Systems

    … and the Multiview Graph Dual Attention and Contrastive Learning for Multi-Criteria Recommender Systems (D-MGAC). The first framework, HFRS-DA, addresses the challenge of effectively integrating heterogeneous information and uncovering meaningful relationships among entities in food and …

    unsw Repository record for Graph Neural Networks for Health-Aware Food and Multi-Criteria Recommendation Systems (opens in a new tab)

  12. Improving utilization, granularity, and interpretability in visual representation learning

    … three works that revolve around improving the learning and usage of deep model features in computer vision. The first work is about improving style transfer, which is a generative artistic method that leverages pretrained deep model features. Style transfer boils down to a distribution matching …

    uiuc Repository record for Improving utilization, granularity, and interpretability in visual representation learning (opens in a new tab)

  13. Computational Methods for Biomedical Imaging

    … by these beams is well-suited for machine learning applications, where 2D images can contain 3D contextual information without the added computational overhead of performing 3D convolutions. We begin with a review of important non-diffracting beams in the existing literature, and proceed to …

    mit Repository record for Computational Methods for Biomedical Imaging (opens in a new tab)

  14. Detecting Irregular Network Activity with Adversarial Learning and Expert Feedback

    … thesis proposes a novel self-supervised deep learning framework CAAD for anomaly detection in wireless communication systems. Specifically, CAAD employs powerful adversarial learning and contrastive learning techniques to learn effective representations of normal and anomalous behavior in …

    vt Repository record for Detecting Irregular Network Activity with Adversarial Learning and Expert Feedback (opens in a new tab)

  15. Contrastive representation learning for bioimage quantification

    Deep learning has enabled unprecedented progress towards automating the analysis and quantification of large-scale, high-resolution imaging data. However, the majority of current deep learning systems for bioimage analysis is trained with manual annotations, leading to limitations in their …

    cambridge Repository record for Contrastive representation learning for bioimage quantification (opens in a new tab)

  16. Hybrid ConVIRT - enhancing medical image-text representation learning of vision language models

    … advancements in image-text representation learning, as demonstrated by Hybrid-ConVIRT, which builds on contrastive learning frameworks such as ConVIRT and MedCLIP. These medical contrastive learning models trained on domain-specific datasets, have tackled issues related to the costly, …

    uoit Repository record for Hybrid ConVIRT - enhancing medical image-text representation learning of vision language models (opens in a new tab)

  17. Robust Learning from Uncurated Data

    The field of machine learning has witnessed a growing interest in learning from uncurated data, which involves training models from data that has not been carefully curated or labeled. However, this type of data is typically noisy, incomplete, and riddled with errors, making it challenging for …

    mit Repository record for Robust Learning from Uncurated Data (opens in a new tab)

  18. Enhancing Self-Supervised Learning through Transformations in Higher Activation Space

    … Augmentation through extensive experiments on contrastive learning tasks in computer vision and NLP domains, where we observe substantial performance gains with ResNets and Transformers as the underlying models. Our experimentation reveals that targeting deeper layers with Deep Augmentation …

    mit Repository record for Enhancing Self-Supervised Learning through Transformations in Higher Activation Space (opens in a new tab)

  19. Classifying and Displaying Brain-waves through Self-supervised Learning

    … constructing labeled datasets for supervised learning from EEG signals is expensive and time-consuming. Moreover, the existing datasets use incompatible EEG setups (e.g. different numbers of channels, sampling rates, types of sensors, etc.) that make them hard to fuse to obtain larger …

    mit Repository record for Classifying and Displaying Brain-waves through Self-supervised Learning (opens in a new tab)

Page 1 of 4