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 8 of 8 for “"kernel alignment"”.

  1. A Kernel-based Approach for Learning Causal Graphs From Mixed Data Containing Missing Values

    … graph is unknown. This research proposes using kernel functions and kernel alignment to handle mixed data. The two main steps of this approach are computing a kernel matrix for each variable and calculating a pseudo-correlation matrix using kernel alignment. The kernel alignment matrix is used …

    whiterose Download PDF for A Kernel-based Approach for Learning Causal Graphs From Mixed Data Containing Missing Values (opens in a new tab)

  2. Kernel Methods in Computer-Aided Constructive Drug Design

    … involve dozens to hundreds of attributes. We use kernel-based methods to define these measures of similarity. Kernels are general functions that can be used to formulate similarity comparisons. The overall goal of this thesis is to develop effective and efficient computational methods that are …

    waterloo Repository record for Kernel Methods in Computer-Aided Constructive Drug Design (opens in a new tab)

  3. Semidefinite Embedding for the Dimensionality Reduction of DNA Microarray Data

    … Our method, <strong><em>KAS</em></strong> (<em>kernel alignment with semidefinite embedding</em>), aids the visualization of microarray data in two dimensions and shows improvement over existing dimensionality reduction methods such as PCA, LLE and Isomap.

    waterloo Repository record for Semidefinite Embedding for the Dimensionality Reduction of DNA Microarray Data (opens in a new tab)

  4. Non-Invasive Human Fatigue Estimation Based on Motion Analysis

    … inertial representations; on WESAD, centered-kernel alignment improves from 0.61 for a supervised baseline to 0.88 for the distilled model, indicating a substantial gain in physiological relevance under single-sensor constraints. The second introduces a lightweight vision-based framework that …

    guelph Repository record for Non-Invasive Human Fatigue Estimation Based on Motion Analysis (opens in a new tab) Embargoed until 2026-12-16

  5. Comparing multimodal and unimodal representation learning for Rare cancer prediction

    … between modalities (measured using Centered Kernel Alignment), but the distinctness of representations does not translate into complementary predictive value for downstream tasks. Our findings establish empirical boundaries on multimodal learning in rare cancer settings characterised by …

    kent Download PDF for Comparing multimodal and unimodal representation learning for Rare cancer prediction (opens in a new tab)

  6. Continual Learning in Constrained Scenarios: Bridging Real-World Needs and Practical Constraints

    … reliability of current metrics, such as Centered Kernel Alignment (CKA), in tracking representation changes and proposes novel methods for more accurate measurement of forgetting. The second area of study involves developing CL methods for restricted scenarios, particularly in cases where …

    concordia Download PDF for Continual Learning in Constrained Scenarios: Bridging Real-World Needs and Practical Constraints (opens in a new tab)

  7. SPARSE REPRESENTATION, DISCRIMINATIVE DICTIONARIES AND PROJECTIONS FOR VISUAL CLASSIFICATION

    … nonlinear extensions of these methods using the kernel trick. We also address the problem of choosing the optimal kernel for sparse representation-based classification using Multiple Kernel Learning (MKL) methods. Finally, in order to deal with heterogeneous multimodal data, we present a feature …

    maryland Repository record for SPARSE REPRESENTATION, DISCRIMINATIVE DICTIONARIES AND PROJECTIONS FOR VISUAL CLASSIFICATION (opens in a new tab)

  8. Training Phase Specific Backdoor Defences For Federated Learning

    The explosion in the availability of data has propelled the use of machine learning across industry and academia. Traditionally, data and computational resources were located on company or institutional premises. However, the current trend sees data scattered across various data silos and devices. …

    unsw Repository record for Training Phase Specific Backdoor Defences For Federated Learning (opens in a new tab)