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 41 for “"Scalable models"”.

  1. Rich and Scalable Models for Text

    Topic models have become essential tools for uncovering hidden structures in big data. However, the most popular topic model algorithm—Latent Dirichlet Allocation (LDA)— and its extensions suffer from sluggish performance on big datasets. Recently, the machine learning community has attacked this …

    maryland Repository record for Rich and Scalable Models for Text (opens in a new tab)

  2. Advances in kernel methods: towards general-purpose and scalable models

    … are presented for constructing fully automated, scalable, and general-purpose kernel machines that perform very well over a wide range of input dimensions and sample sizes. This thesis aims to contribute towards bridging the gap between kernel methods and deep learning and to propose methods that …

    oxford Repository record for Advances in kernel methods: towards general-purpose and scalable models (opens in a new tab)

  3. Flexible and scalable models for clustering and few-shot learning

    … The purpose is to transfer the knowledge from models supervised on large amounts of annotated data to target few-shot tasks, with new (unseen) classes and a few annotated samples per new class. In this thesis, we investigate bound-optimization based clustering methods, which integrate various …

    ets-quebec Download PDF for Flexible and scalable models for clustering and few-shot learning (opens in a new tab)

  4. Navigating Widespread Urban Transit Dynamics with Standardized Data and Scalable Models

    … priority treatments. Then a set of deep learning models are developed to forecast bus travel times under different data availability scenarios. Their generalizability is tested across different cities and transit networks. Finally, these models are used to estimate energy demands of a battery …

    washington Repository record for Navigating Widespread Urban Transit Dynamics with Standardized Data and Scalable Models (opens in a new tab)

  5. Machine learning and causality: Building robust and scalable models for decision-making

    … processes. However, ML models, while powerful, often lack robustness, especially when facing data biases and distribution shifts. Causal methods, focusing on understanding and exploiting causal relationships, offer a promising avenue for building robust models. However, …

    toronto-retro Repository record for Machine learning and causality: Building robust and scalable models for decision-making (opens in a new tab)

  6. Rapid, Predictive Modeling for High Frequency Interconnect on Low Cost Substrates

    In this dissertation, a predictive (scalable) measurement-based PEEC modeling method for high-frequency interconnects on low-cost FR4 substrates is proposed and demonstrated. The interconnects are modeled with equivalent circuits of scalable building blocks using a rapid and accurate optimization …

    gatech Repository record for Rapid, Predictive Modeling for High Frequency Interconnect on Low Cost Substrates (opens in a new tab)

  7. MORPHOLOGICAL CLASSIFICATION OF SUBTYPES OF VOLUMETRIC PROJECTION NEURONS FROM MOUSE BRAIN SCANS

    … study of the brain are in demand. Robust and scalable models for classification of neuron morphologies are necessary for the automation. The goal of this thesis is to introduce a robust and fast methodology based on minute local level features of 3D projection neurons for classification of …

    jhu Repository record for MORPHOLOGICAL CLASSIFICATION OF SUBTYPES OF VOLUMETRIC PROJECTION NEURONS FROM MOUSE BRAIN SCANS (opens in a new tab)

  8. Current and future trends in object detection

    … use of more representative features and better models. We propose new features that are not only more powerful, but also more robust and capture more information than the currently popular features. Further, we propose scalable models which can leverage large amounts of training data to improve …

    uiuc Repository record for Current and future trends in object detection (opens in a new tab)

  9. Deep Learning Approaches to Problems in Speech Recognition, Computational Chemistry, and Natural Language Text Processing

    … to machine learning emphasizes high-capacity, scalable models that learn distributed representations of their input. This dissertation demonstrates the efficacy and generality of this approach in a series of diverse case studies in speech recognition, computational chemistry, and natural …

    toronto-retro Repository record for Deep Learning Approaches to Problems in Speech Recognition, Computational Chemistry, and Natural Language Text Processing (opens in a new tab)

  10. Hybrid prototyping of multicore embedded systems

    … have enabled creation of fast software models of multicore platforms, commonly known as Virtual Prototypes, for early functional validation of embedded software, before the hardware is available. On the other hand, for accurate performance validation, the complete multicore platform can …

    concordia Download PDF for Hybrid prototyping of multicore embedded systems (opens in a new tab)

  11. Event-Aware Imputation and Prediction of Urban Traffic Using Deep Spatiotemporal Learning Models

    … and generative approaches. Random Forest models are employed to provide fast, robust estimates, while Generative Adversarial Imputation Networks (GAIN) refine the results, capturing complex dependencies and uncertainty. Experiments on Hamilton, Ontario data demonstrate that the framework …

    mcmaster Repository record for Event-Aware Imputation and Prediction of Urban Traffic Using Deep Spatiotemporal Learning Models (opens in a new tab)

  12. Implicit Regularization and Inductive Biases in Neural Network Training

    … unprecedented scale.A defining feature of these models is their scalability: as model size, data, and compute increase, performance often improves in a predictable manner, enabling the solution of increasingly complex problems. Despite this success, fundamental questions remain. Neural networks …

    cornell Repository record for Implicit Regularization and Inductive Biases in Neural Network Training (opens in a new tab)

  13. PROPUESTA DE METODOLOGIA PARA LA IDENTIFICACION Y EL ANALISIS DE MODELOS DE NEGOCIO TRIPLEMENTE SOSTENIBLES EN EL SECTOR AGROALIMENTARIO

    … for identifying and analysing business models through case studies, with the objective of creating a triple bottom-line, replicable and non-scalable business model bank in the agri-food industry". The application of this methodology provides a tool that facilitates the replicability of …

    upv Repository record for PROPUESTA DE METODOLOGIA PARA LA IDENTIFICACION Y EL ANALISIS DE MODELOS DE NEGOCIO TRIPLEMENTE SOSTENIBLES EN EL SECTOR AGROALIMENTARIO (opens in a new tab)

  14. Data-Driven Modeling and Experimental Investigation of Anaerobic Digestion of Wastewater Sludge

    … artificial neural networks (ANN), decision-tree models (Random Forest, XGBoost), support vector machines, and K-nearest neighbors, were trained on nearly two decades of published data. ANNs delivered the best performance for solubilization prediction, while decision-tree models excelled in …

    queens Repository record for Data-Driven Modeling and Experimental Investigation of Anaerobic Digestion of Wastewater Sludge (opens in a new tab)

  15. A 40 GHz Power Amplifier Using a Low Cost High Volume 0.15 um Optical Lithography pHEMT Process

    … and the implementation of high frequency scalable models. The three stage power amplifier was designed using the TOM4 scalable depletion mode FET model. The TriQuint TQP15 Design Kit also implements microstrip transmission line models that can be used for evaluating the interconnect lines …

    portland-state Repository record for A 40 GHz Power Amplifier Using a Low Cost High Volume 0.15 um Optical Lithography pHEMT Process (opens in a new tab)

  16. Synergizing Probabilistic Models with Deep Learning: A Novel Framework for Topic Modeling and Multimodal Analysis

    … Within this paradigm, we develop a series of models spanning topic modeling, multimodal fusion, and cross-domain adaptation. We first propose a Generalized Dirichlet Variational Autoencoder (GD-VAE) for neural topic modeling, followed by smoothed Dirichlet-based multimodal architectures for …

    concordia Download PDF for Synergizing Probabilistic Models with Deep Learning: A Novel Framework for Topic Modeling and Multimodal Analysis (opens in a new tab)

  17. Machine learning on building energy modeling : city and building scale

    … applying novel machine learning (ML) surrogate models across city, urban, and building scales. In different scale, data itself present various context with its unique focus and research questions. To best answering these questions, the work devotes to utilize machine learning, a data-driven …

    texas Repository record for Machine learning on building energy modeling : city and building scale (opens in a new tab)

  18. Three Essays in Resort and Sea Cargo Revenue Management

    … This essay also proves that three RLP models are asymptotically optimal and provide a numerical comparison. The second essay contains a data-driven study for resort revenue management. This work describes a complete revenue management system implemented for the golf resort setting with …

    queens Repository record for Three Essays in Resort and Sea Cargo Revenue Management (opens in a new tab)

Page 1 of 3