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 10 of 10 for “"dataset distillation"”.

  1. Unlocking Inexpensive Training Through Overcoming Weaknesses in Dataset Distillation

    Dataset distillation is a nascent technique that promises accelerated neural network training. In particular, distillation produces a highly compressed dataset that can train a neural network in as little as a single optimization step. However, distillation has not been adopted for real-world use …

    byu Repository record for Unlocking Inexpensive Training Through Overcoming Weaknesses in Dataset Distillation (opens in a new tab) Download PDF for Unlocking Inexpensive Training Through Overcoming Weaknesses in Dataset Distillation (opens in a new tab)

  2. TOWARDS DATA-EFFICIENT DEEP LEARNING

    … learning by tackling the limitations of current dataset distillation (DD) methods, which aim to compress large datasets into compact synthetic ones for faster training and enhanced privacy. First, it introduces Dataset Factorization, a novel framework that decomposes datasets into learnable bases …

    nus Repository record for TOWARDS DATA-EFFICIENT DEEP LEARNING (opens in a new tab) Download file for TOWARDS DATA-EFFICIENT DEEP LEARNING (opens in a new tab)

  3. Task Distillation: Transforming Reinforcement Learning into Supervised Learning

    Recent work in dataset distillation focuses on distilling supervised classification datasets into smaller, synthetic supervised datasets in order to reduce per-model costs of training, to provide interpretability, and to anonymize data. Distillation and its benefits can be extended to a wider array …

    byu Repository record for Task Distillation: Transforming Reinforcement Learning into Supervised Learning (opens in a new tab) Download PDF for Task Distillation: Transforming Reinforcement Learning into Supervised Learning (opens in a new tab)

  4. IMPROVED GRADIENT-BASED LEARNING FOR ENHANCED EFFICIENCY AND GENERALIZATION OF DEEP NEURAL NETWORKS

    … tasks including adversarial attacks and dataset distillation. A meta attacker is developed using meta learning for efficient gradient estimation in black-box adversarial attacks. For dataset distillation, the Flat Trajectory for Distillation (FTD) method mitigates accumulated trajectory …

    nus Repository record for IMPROVED GRADIENT-BASED LEARNING FOR ENHANCED EFFICIENCY AND GENERALIZATION OF DEEP NEURAL NETWORKS (opens in a new tab) Download file for IMPROVED GRADIENT-BASED LEARNING FOR ENHANCED EFFICIENCY AND GENERALIZATION OF DEEP NEURAL NETWORKS (opens in a new tab)

  5. Computer Vision in Histopathology Image Analysis: Preprocessing and Classification

    … To find these instances, we introduce a dataset distillation framework tailored to WSI analysis. In the third chapter, we focus on class imbalance learning which is a commonly known issue in WSI datasets. Here, we proposed a method to enhance the feature learning capacity of a …

    unsw Repository record for Computer Vision in Histopathology Image Analysis: Preprocessing and Classification (opens in a new tab)

  6. On the Effect of Data on Image Classification Tasks

    … various domains, but its reliance on large-scale datasets and complex models raises concerns about sustainability and efficiency. This thesis aims to leverage data potential in image classification tasks, focusing on two critical aspects: efficient learning and multi-label representation learning. …

    toronto-retro Repository record for On the Effect of Data on Image Classification Tasks (opens in a new tab)

  7. Personalized Distributed Machine Learning With Graphs

    … data level, we present GSTAM, the first graph-dataset condensation method that aligns structural attention maps of graph neural networks, producing compact yet faithful synthetic graphs that can further be utilized for the designed FL framework and to increase its computation efficiency. …

    toronto-retro Repository record for Personalized Distributed Machine Learning With Graphs (opens in a new tab)

  8. Learning From Almost No Data

    … of training large neural networks on large datasets has provided great returns, but it leaves behind the multitude of researchers, companies, and practitioners who do not have access to sufficient funding, compute power, or volume of data. This thesis aims to rectify this growing imbalance …

    waterloo Repository record for Learning From Almost No Data (opens in a new tab)