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 21 for “"Custom Loss"”.

  1. RIM-Decay: A deep learning method for clinical prediction

    … medical attention would result in the loss of life. Prediction of early mortality, length of stay, and acute respiratory failure based on vital measurements such as temperatures, blood pressure, heart rate, sugar level, etc., would assist in saving life and cost. EHR records are sparse …

    regina Repository record for RIM-Decay: A deep learning method for clinical prediction (opens in a new tab) Download PDF for RIM-Decay: A deep learning method for clinical prediction (opens in a new tab)

  2. DATA-DRIVEN FAILURE PREDICTION FOR MANUFACTURING PROCESSES

    … within the model architecture, while lot-loss L-CNN utilizes a custom loss function for providing run-level predictions. The proposed models are validated on real-world semiconductor industry datasets.

    nus Repository record for DATA-DRIVEN FAILURE PREDICTION FOR MANUFACTURING PROCESSES (opens in a new tab) Download file for DATA-DRIVEN FAILURE PREDICTION FOR MANUFACTURING PROCESSES (opens in a new tab)

  3. Deep learning methods for mitigating catastrophic forgetting in medical imaging

    … By combining momentum contrastive learning and a custom loss function that integrates classification, cosine similarity, and distillation losses, CLFCR-MC enhances the model’s ability to retain previous knowledge while adapting to new tasks. Experiments using medical imaging datasets, such as …

    uvic Repository record for Deep learning methods for mitigating catastrophic forgetting in medical imaging (opens in a new tab)

  4. Designing Customized Loss Functions for Training Deep Neural Networks

    This dissertation explores the critical role of loss functions in enhancing the predictive performance of deep machine learning models. Loss functions are an integral element of all the ongoing advances we witness daily in this domain. I design custom loss functions and their impacts on various …

    denver Repository record for Designing Customized Loss Functions for Training Deep Neural Networks (opens in a new tab)

  5. A Comparative Study of Machine Learning Models for Multivariate NextG Network Traffic Prediction with SLA-based Loss Function

    … models in network traffic prediction using a custom Service-Level Agreement (SLA) - based loss function to ensure SLA violation constraints while minimizing overprovisioning. The proposed SLA-based parametric custom loss functions are used to maintain the SLA violation rate percentages the …

    vt Repository record for A Comparative Study of Machine Learning Models for Multivariate NextG Network Traffic Prediction with SLA-based Loss Function (opens in a new tab)

  6. Low-shot Visual Recognition

    … to obey certain property by using a custom loss function. We believe that when the lowshot sample obey this property the classification step becomes easier. We show that the proposed solution performs better than the softmax classifier by a good margin.

    vt Repository record for Low-shot Visual Recognition (opens in a new tab)

  7. Symbolic Regression for Data-Driven Equation Discovery: A Physics-Informed Approach

    … solutions. We utilize PySR, a highly customizable symbolic regression package, which combines genetic programming and modern optimization methods to efficiently search for interpretable equations. PySR balances model complexity with performance by penalizing overly complex expressions …

    odu Repository record for Symbolic Regression for Data-Driven Equation Discovery: A Physics-Informed Approach (opens in a new tab)

  8. A Machine Learning Model for Trapped-ion State Classification

    … and improved our own FFNN architecture with custom loss functions.

    waterloo Repository record for A Machine Learning Model for Trapped-ion State Classification (opens in a new tab)

  9. Wake modeling of wind turbines based on machine learning methods for wind farm design

    … Navier–Stokes (RANS) equations into a custom loss function. Comparative analyses demonstrate that PI-DeepWFLO achieves improved stability and robustness by reducing sensitivity to variations in network weight initialization and training datasets relative to the purely data-driven …

    toronto-retro Repository record for Wake modeling of wind turbines based on machine learning methods for wind farm design (opens in a new tab)

  10. Methodologies for Systematic Evaluation and Targeted Mitigation of Deficiencies in Critical Machine Learning Models

    … using two complementary strategies: (i) a custom loss function that penalizes violations of medical constraints, and (ii) a rule-based decision tree derived from clinical knowledge, aggregated with a data-driven model. The resulting knowledge-guided models demonstrated notable improvements …

    vt Repository record for Methodologies for Systematic Evaluation and Targeted Mitigation of Deficiencies in Critical Machine Learning Models (opens in a new tab)

  11. Multi-channel swin transformer framework for rolling bearing remaining useful life prediction

    … of degradation. Additionally, a safety-aware loss function is introduced to prioritize safety by penalizing late predictions more heavily than early ones, aligning the learning process with the asymmetric risk profile of industrial failure scenarios. The framework is evaluated on the PRONOSTIA …

    uvic Repository record for Multi-channel swin transformer framework for rolling bearing remaining useful life prediction (opens in a new tab)

  12. Application of Machine Learning Techniques to Forecast Harmful Algal Blooms in Gulf of Mexico

    … in the grid within each data point, we use a custom loss function that ignores prediction errors on missing cells. Thus the loss function critiques the models based on known cells alone, while the models act with (forward/backward) predictions that are spatiotemporally consistent across both …

    usm Repository record for Application of Machine Learning Techniques to Forecast Harmful Algal Blooms in Gulf of Mexico (opens in a new tab)

  13. Achieving Security and Reliability of Industrial Control Systems Using Data-Driven Models Informed by Physical Domain Knowledge

    … graph neural network, trained using a custom loss function defined by the power flow model in addition to training data, to quickly localize attacks and adapt to different topologies. By embedding knowledge of the physics of the grid, the resulting data-driven model is able to transfer …

    gatech Repository record for Achieving Security and Reliability of Industrial Control Systems Using Data-Driven Models Informed by Physical Domain Knowledge (opens in a new tab)

  14. BEYOND FINITE ELEMENT : PHYSICS INFORMED NEURAL NETWORK FOR STRESS PREDICTION

    … damaged steel plates using CNNs augmented with custom loss functions that use physics rules to bypass the need for Finite Element Analysis. We embedded physics constraints into the loss function to enforce the model training, precisely capturing stress concentrations around the tips of various …

    msu Repository record for BEYOND FINITE ELEMENT : PHYSICS INFORMED NEURAL NETWORK FOR STRESS PREDICTION (opens in a new tab)

  15. Addressing Data Scarcity with Computer Vision Methods

    … deep learning systems. We propose a novel custom loss function, Dense Loss, which was designed to use local region homogeneity regularization to promote contiguous and smooth segmentation predictions while also using an L1-norm loss to reconstruct dense-labelled annotation ground truth for …

    uvic Repository record for Addressing Data Scarcity with Computer Vision Methods (opens in a new tab)

  16. Learned Descriptors for Scalable and Efficient Visual Place Recognition

    … data subsets, a modified architecture, and a custom loss function for stable contrastive training, optimised to generate robust learned image representations. Thirdly, we propose an efficient scene categorisation method leveraging Variational Autoencoders (VAEs). Our approach encodes images …

    maynooth Download PDF for Learned Descriptors for Scalable and Efficient Visual Place Recognition (opens in a new tab)

  17. Time-to-Event Prediction Using Deep Learning Models: Application to GPU Failure Data

    … vectors. We utilize the Cross-Entropy Loss function to optimize TypeEmbedNet and the Mean Squared Error (MSE) for TimeEmbedNet. We develop evaluation metrics to comprehensively assess the models' performance in predicting both failure time and status, taking into account the …

    vt Repository record for Time-to-Event Prediction Using Deep Learning Models: Application to GPU Failure Data (opens in a new tab)

  18. sEMG-based Predictive Framework for Injury Risks in Sports: Novel Deep Learning Approaches Tested on EPL Athletes

    … The drill-time labelling model incorporates custom loss regularisation specific to football training and a kernel and depth selector attention module. This module enhances feature learning versatility and model performance, accurately identifying drill times amid diverse athletic …

    kings Repository record for sEMG-based Predictive Framework for Injury Risks in Sports: Novel Deep Learning Approaches Tested on EPL Athletes (opens in a new tab) Download PDF for sEMG-based Predictive Framework for Injury Risks in Sports: Novel Deep Learning Approaches Tested on EPL Athletes (opens in a new tab)

  19. From surface to tropopause: on the vertical structure of the tropical cyclone vortex

    … to train a neural network (NN) to predict WRs. A custom loss function ensures the model prioritizes accurate prediction of the strongest wind observations which are uncommon. The NN can leverage relevant physical relationships from the observational data and predict a surface wind field in …

    colostate Repository record for From surface to tropopause: on the vertical structure of the tropical cyclone vortex (opens in a new tab)

  20. Artificial neural networks for problems in computational cognition

    … deep learning attention architecture, and a custom loss function, and compare the performance of our model to a number of non-neural network approach baselines. I find that all versions of our model significantly outperform the baseline approaches, and that our novel loss improves on …

    glasgow Repository record for Artificial neural networks for problems in computational cognition (opens in a new tab) Download PDF for Artificial neural networks for problems in computational cognition (opens in a new tab)

Page 1 of 2