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 13 of 13 for “"Dynamic pruning"”.

  1. Communication-efficient personalization in federated learning for edge devices

    … data. First, we introduce a dataset-aware dynamic pruning strategy coupled with gradient control to curb overfitting on heterogeneous clients, stabilize convergence, and lower both computation and communication during local updates. Next, we propose a multimodal federated framework with …

    iastate Repository record for Communication-efficient personalization in federated learning for edge devices (opens in a new tab)

  2. Collaborative Distributed Machine Learning: from knowledge reuse to sparsification in federated learning

    … distributed (non-i.i.d.), we incorporate pruning techniques to enhance both system efficiency and effectiveness. This approach reduces communication and computation costs while simultaneously improving model performance and accuracy. Centralized federated learning (CFL) relies on a central …

    glasgow Repository record for Collaborative Distributed Machine Learning: from knowledge reuse to sparsification in federated learning (opens in a new tab) Download PDF for Collaborative Distributed Machine Learning: from knowledge reuse to sparsification in federated learning (opens in a new tab)

  3. Intelligent control system for CFD modelling software

    … an additional expert system responsible for dynamic pruning of the search space using the rules obtained by statistical analysis of the initial results. Other features include dedicated goal-driven search plans that help reduce the search space even further. The simulation results and overall …

    greenwich Download PDF for Intelligent control system for CFD modelling software (opens in a new tab)

  4. Software-Hardware Co-Design of Efficient Rendering and Processing for Modern AR/VR

    … to incorporate geometric properties and dynamic pruning, optimizing SR to achieve high-quality results with reduced computational overhead. Our SR techniques enhance the efficiency of rendering processes, making them more suitable for real-time applications. Our comprehensive evaluations …

    houston Repository record for Software-Hardware Co-Design of Efficient Rendering and Processing for Modern AR/VR (opens in a new tab)

  5. Learning Neural Network Architecture from Data: NAS and Dynamic Networks

    … data, e.g. neural architecture search (NAS) and dynamic network, are novel ways of architecture design. These newly rising data-dependent methods have great potential but also have many fatal problems yet to be solved. On one hand, the efficiency and effectiveness of NAS cannot be guaranteed at …

    uts Repository record for Learning Neural Network Architecture from Data: NAS and Dynamic Networks (opens in a new tab)

  6. Deep learning and its applications in wireless big data networks performance

    … and training time are presented, such as dynamic pruning during the training phase, based on criteria inspired by network theory (e.g. scale free networks). Moreover, we introduce algorithms leveraging pre-trained neural network weights (transfer learning), accelerating convergence and …

    greece Repository record for Deep learning and its applications in wireless big data networks performance (opens in a new tab)

  7. Effective and efficient transformer models for sequential recommendation

    … of documents has been addressed using Dynamic Pruning approaches that allow finding Top-K items without scoring the whole catalogue exhaustively. Building upon the similarities between item representations in RecJPQ and document representations in IR, we propose the RecJPQPrune dynamic …

    glasgow Repository record for Effective and efficient transformer models for sequential recommendation (opens in a new tab) Download PDF for Effective and efficient transformer models for sequential recommendation (opens in a new tab)

  8. Efficient processing of advanced structural queries in graph databases

    … constraint for candidate generation. A dynamic pruning technique and an improved verification algorithm are also developed. In addition, a cost-aware graph partitioning technique is conceived to optimize the index. Equipping structural queries with rich semantics, substructure selection …

    unsw Repository record for Efficient processing of advanced structural queries in graph databases (opens in a new tab)

  9. An Information-Theoretic Approach To Practical Side Channel Protection

    … Finally, we study inference-phase attacks on dynamically pruned convolutional neural networks (CNNs), utilizing our theoretical results. Dynamic pruning is an increasingly popular technique for optimizing the execution time of the inference phase. We show that these techniques create a highly …

    cornell Repository record for An Information-Theoretic Approach To Practical Side Channel Protection (opens in a new tab)

  10. Computational Architecture for Intelligent Edge Computing

    … techniques, particularly quantization and pruning, are therefore essential enablers of edge AI, as they reduce numerical precision and remove redundant parameters to lower computational and memory costs. However, existing approaches suffer from two key limitations: aggressive compression …

    auckland-ms Repository record for Computational Architecture for Intelligent Edge Computing (opens in a new tab)

  11. Effective Math-Aware Ad-Hoc Retrieval based on Structure Search and Semantic Similarities

    … efficient structure search, I introduce a dynamic pruning algorithm to the problem of structure retrieval. The proposed retrieval algorithm efficiently identifies the maximum common subtree among formula candidates and safely eliminates potential structure matches that exceed a dynamic …

    waterloo Repository record for Effective Math-Aware Ad-Hoc Retrieval based on Structure Search and Semantic Similarities (opens in a new tab)

  12. Efficient, Robust, and Self-Supervised Algorithm-Hardware Co-Design

    … point, this research presents a structured pruning algorithm followed by the dedicated in-memory-computing deployment scheme. While the efficiency has been improved, the non-ideality of the individual ReRAM device and non-reliable retention characteristics destroy the accuracy of the …

    cornell Repository record for Efficient, Robust, and Self-Supervised Algorithm-Hardware Co-Design (opens in a new tab)

  13. Υπολογιστική νοημοσύνη - διαχείριση οπτικής πληροφορίας

    … research work, a method was proposed for the dynamic optimization, in terms of energy consumption and execution speed, of pre-trained CNNs, with appropriate transformations in their structure. In the generated dynamic CNNs, only the kernels which are useful for processing the given input data …

    greece Repository record for Υπολογιστική νοημοσύνη - διαχείριση οπτικής πληροφορίας (opens in a new tab)