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 155 for “"Catastrophic Forgetting"”.

  1. Catastrophic Forgetting in Language Models

    Catastrophic forgetting remains a persistent challenge in the continual learning paradigm of neural networks, particularly in the context of pre-trained language models. This thesis investigates the phenomenon of catastrophic forgetting in large language models (LLMs), with a focus on BERT, through …

    chalmers Repository record for Catastrophic Forgetting in Language Models (opens in a new tab)

  2. Overcoming catastrophic forgetting in heterogeneous federated learning

    … patient privacy. However, key challenges such as catastrophic forgetting and local overfitting hinder the effectiveness of FL models in heterogeneous healthcare environments. This thesis addresses these challenges through two complementary studies. The first study introduces Federated Impressions, …

    ubc Repository record for Overcoming catastrophic forgetting in heterogeneous federated learning (opens in a new tab) Download PDF for Overcoming catastrophic forgetting in heterogeneous federated learning (opens in a new tab)

  3. Reducing Catastrophic Forgetting in Self-Organizing Maps

    … from new data samples. This problem is known as catastrophic forgetting and remains an unsolved problem in the domain of machine learning to this day. To overcome catastrophic forgetting, different approaches have been proposed. One major line of thought advocates the use of memory buffers to …

    rit Repository record for Reducing Catastrophic Forgetting in Self-Organizing Maps (opens in a new tab)

  4. New Insights on Catastrophic Forgetting in Neural Networks

    … adaptation of new information while combating catastrophic forgetting, \textit{i.e.} stability-plasticity dilemma. This thesis is comprised of four major chapters that explore different aspects of continual learning and propose novel solutions to address some of its challenges. First, we …

    concordia Download PDF for New Insights on Catastrophic Forgetting in Neural Networks (opens in a new tab)

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

    … time. A persistent challenge in this field is catastrophic forgetting, where models overwrite past knowledge when learning new tasks, limiting their practical use in dynamic environments. This thesis introduces a new framework called CLFCR-MC (Continual Learning Framework with Contrastive …

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

  6. Neuron Clustering for Mitigating Catastrophic Forgetting in Supervised and Reinforcement Learning

    … in more realistic online environments is that of catastrophic forgetting. Catastrophic forgetting affects supervised learning systems when input samples are temporally correlated or are non-stationary. However, most real-world problems are non-stationary in nature, resulting in prolonged periods …

    utk-theses Repository record for Neuron Clustering for Mitigating Catastrophic Forgetting in Supervised and Reinforcement Learning (opens in a new tab)

  7. Referencing Unlabelled World Data to Prevent Catastrophic Forgetting in Class-incremental Learning

    … a novel strategy to address the challenge of "catastrophic forgetting" in deep continual-learning systems. The term refers to severe performance degradation for older tasks, as a system learns new tasks that are presented sequentially. Most previous techniques have emphasized preservation of …

    vt Repository record for Referencing Unlabelled World Data to Prevent Catastrophic Forgetting in Class-incremental Learning (opens in a new tab)

  8. Beyond catastrophic forgetting : advancing continual learning for robust and fair medical image analysis

    … models to learn sequentially while addressing catastrophic forgetting. However, challenges remain in resource efficiency, generalization, adaptation, fairness and the inefficient integration of pretrained models, especially in medical image analysis. To tackle these challenges, this …

    ubc Repository record for Beyond catastrophic forgetting : advancing continual learning for robust and fair medical image analysis (opens in a new tab) Download PDF for Beyond catastrophic forgetting : advancing continual learning for robust and fair medical image analysis (opens in a new tab)

  9. The Impact Of Experience Replay Buffers On Catastrophic Forgetting Within Proximal Policy Optimization

    Catastrophic Forgetting continues to be a prevalent issue for any neural-based approach to learning. Without better understanding and solutions to solving catastrophic forgetting, learning performance and efficacy in autonomous agents, particularly deep learning-based agents will become drastically …

    ucf Repository record for The Impact Of Experience Replay Buffers On Catastrophic Forgetting Within Proximal Policy Optimization (opens in a new tab)

  10. Developing Generative Adversarial Networks for Classification and Clustering: Overcoming Class Imbalance and Catastrophic Forgetting

    … network is introduced to handle the catastrophic forgetting of past tasks. As the job grows, the network structure of ACL also grows simultaneously. Therefore, ACL is not scalable. We propose an online Scalable Adversarial Continual Learning (SCALE) framework where the parameter …

    unsw Repository record for Developing Generative Adversarial Networks for Classification and Clustering: Overcoming Class Imbalance and Catastrophic Forgetting (opens in a new tab)

  11. Learning with Distribution Shift: Theory and Practice

    … in static environments, they suffer from \emph{catastrophic forgetting} in CL scenarios - a phenomenon where performance on earlier tasks deteriorates after learning new ones. Although significant efforts have been devoted to developing practical CL algorithms, the theoretical understanding of …

    rice Repository record for Learning with Distribution Shift: Theory and Practice (opens in a new tab)

  12. Brain-Inspired Continual Learning: Rethinking the Role of Features in the Stability-Plasticity Dilemma

    … existing approaches to CL face the challenge of catastrophic forgetting, which occurs when a model's performance on previously learned tasks declines after learning the new task. In this dissertation, we focus on the crucial role of input data features in determining the robustness of CL models …

    rowan Repository record for Brain-Inspired Continual Learning: Rethinking the Role of Features in the Stability-Plasticity Dilemma (opens in a new tab)

  13. Deep Convolutional Spiking Neural Networks for Image Classification

    … our network are also explored. We also studied catastrophic forgetting and its effect on spiking neural networks (SNNs). For the experiments regarding catastrophic forgetting, in the classification sections of the network we use a modified synaptic intelligence that we refer to as cost per …

    boise Repository record for Deep Convolutional Spiking Neural Networks for Image Classification (opens in a new tab)

  14. Fast and Sustainable Video Anomaly Detection with Continual Learning

    … under these conditions and may suffer catastrophic forgetting when adapting to new environments. This thesis proposes RegiGrow, a parameter-efficient continual adaptation framework built on the Flashback retrieval pipeline. RegiGrow integrates Mixture-of-Experts Low-Rank Adaptation into …

    sdstate Repository record for Fast and Sustainable Video Anomaly Detection with Continual Learning (opens in a new tab)

  15. Advanced continual machine learning

    Deep learning suffers from catastrophic forgetting when learning a dynamic environment, where the data is continuously gathered. This thesis is proposed to tackle the catastrophic forgetting problem under various settings and constraints, i.e., rehearsal-based, rehearsal-free, data scarcity, …

    adelaide Repository record for Advanced continual machine learning (opens in a new tab)

  16. Multi-Task Reinforcement Learning: From Single-Agent to Multi-Agent Systems

    … multi-task learning due to a phenomenon known as catastrophic forgetting. In this thesis, we present a set of simulation environments for exploring the abilities of multi-task drone systems and provide a platform for testing agents in incremental single-agent and multi-agent learning scenarios. …

    vt Repository record for Multi-Task Reinforcement Learning: From Single-Agent to Multi-Agent Systems (opens in a new tab)

  17. Mitigation of Catastrophic Interference in Neural Networks and Ensembles using a Fixed Expansion Layer

    <p>Catastrophic forgetting (also known in the literature as catastrophic interference) is the phenomenon by which learning systems exhibit a severe exponential loss of learned information when exposed to relatively small amounts of new training data. This loss of information is not caused by …

    utk-theses Repository record for Mitigation of Catastrophic Interference in Neural Networks and Ensembles using a Fixed Expansion Layer (opens in a new tab)

  18. Lifelong Machine Learning With Adaptive Resonance Theory

    … with the mitigation of the phenomenon of catastrophic forgetting whereby learning agents that are faced with incrementally novel information deleteriously overwrite previous knowledge if that learning process is not regularized to counteract this consequence. ART algorithms solve this …

    must-thes Repository record for Lifelong Machine Learning With Adaptive Resonance Theory (opens in a new tab)

  19. Towards efficient deep learning for computer vision

    … issue with deep neural networks is catastrophic forgetting when encountering new or evolving tasks in a sequential manner. The model must be retrained with all the data or tasks encountered to avoid forgetting, thus making them unsuitable for many real-world applications. The second …

    freiburg-diss Repository record for Towards efficient deep learning for computer vision (opens in a new tab)

Page 1 of 8