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 15 of 15 for “"Sparse neural networks"”.

  1. Toward efficient deep learning with sparse neural networks

    … the challenge, this thesis focuses on studying sparse neural networks, particularly around their construction, initialization, and large-scale training aspects, as an attempt to take a step toward efficient deep learning.</p> <p>Firstly, this thesis addresses the problem of finding sparse neural …

    oxford Repository record for Toward efficient deep learning with sparse neural networks (opens in a new tab)

  2. Sparsity prior in efficient deep learning based solvers and models

    … parameterizations by absorbing insights from the sparse optimization field, which result in compact deep-learning-based solvers with significantly reduced training costs but superior empirical performance. In the second direction, we introduce sparsity to deep neural networks via weight pruning. …

    texas Repository record for Sparsity prior in efficient deep learning based solvers and models (opens in a new tab)

  3. Efficient Sparse Neural Network Training

    Developments in neural networks have led to advanced models requiring large amounts of training time and resources. To reduce the environmental impact and to decrease the training times of models, acceleration techniques have been developed. One method is neural network pruning, which removes …

    oregon Repository record for Efficient Sparse Neural Network Training (opens in a new tab)

  4. Chasing sparsity : from model, to algorithm, to science

    … to model the underlying low dimensionality of neural networks, for understanding their generalization, optimization dynamics, implicit regularization, expressivity, and robustness. Meanwhile, appropriate sparsity-aware priors assist deep neural networks to achieve significantly enhanced …

    texas Repository record for Chasing sparsity : from model, to algorithm, to science (opens in a new tab)

  5. Causal Inference with Selection Bias and Complex Observation Data

    … generating distributions. The approach leverages sparse neural networks for their comparable ability to learn complex relationships and implicitly handle variable interactions to that of regular neural networks, while simultaneously aiming for explainability by including feature sparsity. Third, …

    buffalo Repository record for Causal Inference with Selection Bias and Complex Observation Data (opens in a new tab)

  6. Towards fair and efficient distributed intelligence

    … to enhance our world through the use of Deep Neural Networks (DNNs). However, modern DNNs suffer from high computational complexity and are often relegated to specialized hardware, a bottleneck which has severely limited their practical use. In this work, we contribute to improving these …

    colostate Repository record for Towards fair and efficient distributed intelligence (opens in a new tab)

  7. Leveraging sparsity in deep neural networks for training efficiency, interpretability and generalization

    Sparse neural networks (Sparse NNs) are characterized by having fewer connections between consecutive layers compared to traditional fully connected, or dense NNs. Historically, sparsity has been studied post-training to enhance inference efficiency and as a regularization mechanism to improve …

    gatech Repository record for Leveraging sparsity in deep neural networks for training efficiency, interpretability and generalization (opens in a new tab)

  8. On explainability of neural networks

    <p>It is widely reported that deep neural networks outperform most competitors for a range of applications. The state-of-the-art neural networks have built-in inductive bias of architectural choices, regularizations, optimizer types, and initialization methods. Using inductive bias is intuitive to …

    njit Repository record for On explainability of neural networks (opens in a new tab)

  9. Overcoming Noise and Variations In Low-Precision Neural Networks

    … and training choices on the resilience of a neural network when subjected to noise and/or device variations. Simulations were performed under the expectation that the neural network would be implemented on analog hardware; this context asserts that there will be random noise within the …

    gatech Repository record for Overcoming Noise and Variations In Low-Precision Neural Networks (opens in a new tab)