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 23 for “"Regularization techniques"”.
-
Higher-order Beam Theories for Continuum Damage Mechanics with Regularization Techniques
L'abstract è presente nell'allegato / the abstract is in the attachment
-
High resolution signal and image recovery: Fast algorithms and analysis
… signal recovery algorithms, and developing new regularization techniques for reducing the ill effects of noise in signal recovery algorithms.
-
Studies of Atmospheric Gravity Waves in the Mesopause Region Using Airglow Imaging
… from multisensor imaging measurements. Regularization techniques have been compared in the tomographic reconstruction processes of retrieving synthetic wave structures. Various combinations of number and separation of imagers are considered for showing the effect of system geometry on …
-
Sparse seismic signal processing using adaptive dictionaries
… FWI is inherently a challenging problem, so that regularization techniques are typically applied to yield better posed models. Moreover, FWI also suffers from its prohibitive computational costs that mainly arise from forward modeling of the seismic wavefield for multiple sources at each iteration …
-
Embedding and latent variable models using maximal correlation
… pairs that skew the result. We derive simple regularization techniques to compensate for those outliers. Additionally, optimizing for the preservation of maximal correlations after processing lets us induce informative soft clustering and mixture models. Empirical results on natural language …
-
Learning Hyperparameters for Inverse Problems by Deep Neural Networks
… in a process that is often referred to as regularization. Most regularization techniques require suitable choices of regularization parameters. In this work, we will describe new approaches that use deep neural networks (DNN) to estimate these regularization parameters. We will train …
-
Source distribution analysis of magnetic microscopy maps of geological samples
… these problems, we have implemented several regularization techniques and constraints. Using synthetic, computationally generated measurements, our investigation demonstrates that Tikhonov regularization with a high pass filter matrix performs better than unregularized least square methods, …
-
Tomographic reconstruction with adaptive sparsifying transforms
… has been shown to outperform traditional regularization techniques. However, these methods scale poorly with data size, and may be prohibitively expensive for practical tomographic reconstruction. In this thesis, we propose a new method for image reconstruction from low-dose data. The …
-
On the Use of Arnoldi and Golub-Kahan Bases to Solve Nonsymmetric Ill-Posed Inverse Problems
… iterations before solutions become contaminated. Regularization methods such as spectral filtering methods use the singular value decomposition (SVD) and are effective at filtering inverted noise from solutions, but are computationally prohibitive on large problems. Hybrid methods apply …
-
ModelPred: A Framework for Predicting Trained Model from Training Data
… parameters. We introduce novel global and local regularization techniques to prevent overfitting and we rigorously characterize the expressive power of neural networks (NN) in approximating the end-to-end training process. Through extensive empirical investigations, we show that ModelPred enables …
-
A Unified Robust Minimax Framework for Regularized Learning Problems
Regularization techniques have become a principled tool for model-based statistics and artificial intelligence research. However, in most situations, these regularization terms are not well interpreted, especially on how they are related to the loss function and data matrix in a given statistic …
-
Interpretability by Design: New Interpretable Machine Learning Models and Methods
… that, by designing novel model architectures or regularization techniques, we can build machine learning models that are both accurate and interpretable.</p>
-
Fractals in mechanics of materials
… microstructured materials. Using dimensional regularization techniques, a fractional integral is introduced to reflect the mass scaling on fractals. We propose a product measure consistent with generally anisotropic fractals and also simplify previous formulations from decoupling of coordinate …
-
Systems Pharmacology – Machine Learning Approaches in Profiling Oncology Drug Candidates
… (CART) and multi-tree majority voting ensemble techniques i.e., random forest and XGBoost.The feature sets for building these models were extracted by computing chemical fingerprints and quantum chemical descriptors. We generated both sparse and dense matrices for modeling. We cross-validated, …
-
Computational Solution Of Inverse Problems With Simulated Annealing
… and accurately by various numerical solution techniques. The cost function, however, is typically a complicated multidimensional, multimodal surface. These properties make it difficult to locate the global minimum where the quasi-solution exists.;The simulated annealing algorithm performs well …
-
Learning Generative Models Using Structured Latent Variables
… on large labeled datasets, leveraging effective regularization techniques and architectural design. Using more data and computational resources, performance is likely to continue to improve in the future. Despite these nice properties, supervised neural networks are sometimes criticized because …
-
Structure Borne Noise Analysis Using Helmholtz Equation Least Squares Based Forced Vibro Acoustic Components
… were taken to achieve the goal. First, hybrid regularization techniques were developed to improve the reconstruction accuracy of normal surface velocity of the original HELS method. Second, correlations between the surface vibro-acoustic responses and acoustic radiation were factorized using …
-
Improving problem-solving capabilities of language model: data, architecture and algorithms
… followed by the application of projection techniques to generate new textual inputs. This augmentation significantly elevates the performance of knowledge distillation from teacher models, consequently enhancing the capabilities of student models in addressing closed-domain challenges, …
-
Utilizing an ExB probe for obtaining ion velocity distribution function for an electric propulsion system
… that eliminates the need for conventional regularization techniques. A rigorous uncertainty propagation methodology is derived to quantify how measurement errors influence the reconstructed IVDF, thereby enabling statistically robust interpretation of experimental data. The proposed …
-
Multidimensional and High Frequency Heat Flux Reconstruction Applied to Hypersonic Transitional Flows
… in the short time scale, that the L-curve regularization needed to be locally corrected to analyze transitional flows and that proper regularization led to sub-cell resolution of the inverse problem. While the L2 regularization techniques are accurate they are also computationally …
Page 1 of 2