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 38 for “"Model Aggregation"”.
-
Analysis of Model Aggregation Techniques in Federated Learning
… distributed machine learning method that enables model training on client devices, which protects the privacy of user data. In federated learning, instead of passing clients’ data to a server, clients generated models are transferred to the central server, which in turn aggregates all models into …
-
Federated Learning Model Aggregation in Heterogeneous Aerial and Space Networks
… convergence. This algorithm employs weighted aggregation and client selection based on an estimated diversity measure, termed projection, enabling WeiAvgCS to outperform other benchmarks without compromising privacy. For high-degree label heterogeneity, the FedBalance algorithm is proposed, …
-
Adaptive Model Aggregation for Decentralized Federated Learning in Vehicular Networks
… collaborative development of machine learning models, offering a unique balance between collaborative learning and individual data privacy. This approach is particularly apt in scenarios where vehicles, equipped with their own data and computation capabilities, work together to build a …
-
A Secure and Explainable AI-Based Framework for IIoT with Privacy-Prioritized Model Aggregation
… and industrial systems. Beyond merely fortifying model robustness against adversarial attacks, the proposed frameworks are characterized by a steadfast commitment to prioritizing privacy. Moreover, it furnishes effective tools for attack attribution and anomaly prediction within dynamic and …
-
Improved and domain specific model soups
This thesis investigates advanced model aggregation techniques, specifically Model Soups, to enhance the performance and efficiency of Graph Neural Networks (GNNs). Traditional ensemble methods, while accurate, incur significant computational and memory costs, limiting scalability. We propose two …
-
Bayesian analysis of rainfall-runoff models: insights to parameter estimation, model comparison and hierarchical model development
… Estimation of parameter uncertainty (and model uncertainty) allows assessment of the risk in likely applications of hydrological models. Bayesian statistical inference, with computations carried out via Markov Chain Monte Carlo (MCMC) methods, offers an attractive approach to model …
-
Evolutionary Weights for Random Subspace Learning
… Data Mining, this method allows us to aggregate models to reduce prediction error. There are many methods on how to perform model aggregation, one of them is known as Random Subspace Learning, which consists of building subspace of the feature space where we want to create our models. The task of …
-
DATA PROTECTION IN FEDERATED LEARNING
… fairness and data privacy. Fairness ensures models do not favor specific groups, while privacy protects sensitive training data. These challenges become more complex in distributed settings, where data is spread across multiple entities (e.g., edge devices, governments, and service …
-
Proteostasis Responses to Endogenous Alpha-Synuclein Aggregation in the Brain
α-Synuclein aggregation is implicated in several neurodegenerative diseases, including Parkinson’s disease (PD) and dementia with Lewy Bodies (DLB). Changes in cellular signaling pathways induced by this aggregation may contribute to cell death and disease pathogenesis. To investigate this, we used …
-
Hierarchical Federated Learning for Predictive, Scalable, and Privacy-Preserving Service Function Chaining in Cloud-Native Networks
… distributed demand forecasting, federated model aggregation, and SDN-based flow control within a hierarchical orchestration framework. At the edge, HiFL Agents employ Light Gradient-Boosting Machine (LGBM) models to forecast local CPU and memory demand, while the central HiFL Server …
-
GraphDHT: Scaling Graph Neural Networks' Distributed Training on Edge Devices on a Peer-to-Peer Distributed Hash Table Network
… core is weighted graph partitioning and model aggregation in GNNs, based on partition ratios, improving training efficiency and resource use. ALBP promotes inclusive device participation in training, overcoming computational limits and privacy concerns in large-scale graph data …
-
COLLABORATIVE DISTRIBUTED DEEP LEARNING SYSTEMS ON THE EDGES
… compromising the accuracies of the trained models. The third system dubbed FedAce is a distributed training system that improves training efficiency under federated learning setting where private on-device data are not allowed to be shared among local devices. Within each participating …
-
Federated learning model complexity vs robustness to non-IID data and selective federated learning
Federated learning trains a global model using data distributed across local nodes, and differs from centralized machine learning by moving the computation to the data in order to address the challenges of data ownership, privacy, computational power, and data storage. Previous federated learning …
-
Equivalence of Queries with Nested Aggregation
… between conjunctive SQL queries containing aggregation operators that may be nested. Our focus is on understanding the interaction between nested aggregation operators and the other parts of the query body, and so we model aggregation functions simply as abstract collection constructors. …
-
Federated Learning and Meta-Learning for Medical Image Classification: Effective Aggregation Techniques and Adaptive Learning Rate Approaches
… of learning methods: federated learning (FL) and model-agnostic meta-learning (MAML). In the first learning method, FL enables CAD systems to train on diverse data across multiple institutions, overcoming individual dataset limitations. However, data heterogeneity (e.g., due to non-independent and …
-
TRANSFER LEARNING IN CLASSIFICATION AND REGRESSION WITH SUMMARY STATISTICS
… In addition, an adaptive method together with model aggregation is introduced that identifies the possible informative data sets in transfer learning. We show that under some assumptions, Trans-LDA has smaller error rate in estimating the discriminant direction, and smaller classification …
-
Data Security and Privacy Preservation for Mobile Edge Computing
… SVTAgg, a secure, verifiable, and traceable model aggregation protocol for MEC in federated learning based data processing. Unlike conventional federated learning architectures that assume trustworthy aggregation servers, SVTAgg explicitly addresses the privacy and integrity risks posed by …
-
From Enormous Structured Models to On-device Federated Learning: Robustness, Heterogeneity and Optimization
… (2) an explosion of scale, where increasing model size consistently improves performance, and (3) federated learning, where model training moves to mobile devices, where the data is generated and resides. This dissertation presents diagnostic methods and new algorithms to measure and improve …
-
Knowledge Distillation and Its Application to Network Traffic Classification
… the performance of a low-complexity student model with the help of a more powerful teacher. In this dissertation, we study knowledge distillation and its application to network traffic. We posit that knowledge distillation is an essential ingredient of traffic classification. At a network …
-
Privacy Preservation and Verifiability for Federated Learning
… leakage, which allows training a learning model using distributedly stored data without exposing them. In federated learning, multiple clients collaboratively train a single global model that is improved iteratively through clients' local data. Each client receives the global model from an …
Page 1 of 2