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 “"Model generalization"”.
-
Machine Learning Beyond Accuracy: A Features Perspective On Model Generalization
… does not convey the full picture. Existing ML models turn out to be remarkably brittle: a striking example of which is their susceptibility to imperceptible input perturbations known as adversarial examples. In the first part of this thesis, we revisit adversarial examples, to use them as a …
-
A model generalization study in localizing indoor cows with cow localization (colo) dataset
… task difficult, such as the scarcity of data for model fine-tuning and the inability to generalize models effectively. To address these challenges, we introduces COLO (COw LOcalization), a publicly available dataset comprising localization data for Jersey and Holstein cows under various lighting …
-
Improving Model Generalization of Pneumonia Detection from Chest Xray Images Using Deep Learning and Transfer Learning
… tool. Despite their promise, deep learning models for pneumonia detection often face limitations in generalization, performing strongly on familiar datasets but losing accuracy when applied to unseen data from different clinical environments. This study addresses that challenge by applying …
-
Learning-Based Complex Terrain Navigation Under Uncertainty
… mitigating risk due to uncertainty in learned models and improving model generalization in novel environments. To address these challenges, this thesis presents a unified framework to learn uncertainty-aware, physics-informed traversability models and achieve risk-aware navigation in both …
-
On Neural Network Pruning’s Effect on Generalization
… frequently observe that pruning improves model generalization. A longstanding hypothesis attributes such improvement to model size reduction. However, recent studies on over-parameterization characterize a new model size regime, in which larger models achieve better generalization. A …
-
TOWARDS RELIABLE AI UNDER DISTRIBUTION SHIFTS: A DATA-CENTRIC PERSPECTIVE
Machine learning (ML) models often rely on spurious correlations in the training data, leading to performance degradation and unreliability when processing inputs under distribution shifts. This thesis systematically studies the robustness to distribution shifts for ML models from a data-centric …
-
Physics-Informed Deep Learning for Plasma Etch Optimization
Modeling the plasma etch process is highly valuable in the field of semiconductor manufacturing. This intricate process relies on the execution of numerous individual processes, numbering in the hundreds to thousands. The application of artificial intelligence (AI) and machine learning (ML) …
-
Novel Instance-Level Weighted Loss Function for Imbalanced Learning
… classes, which almost never exists in real-world modeling. In the imbalanced data setting, the equal class distribution is grossly violated, and the resulting parameter estimates are biased toward the majority class. To overcome the bias and improve model generalization, we focus on modifying the …
-
Investigation of machine learning techniques in stabilization of co-propagating polarization encoded photons
… we analyze the performance of attention-based models, sliding window time series predictors, and reinforcement learning frameworks in predicting and stabilizing polarization states. Our findings indicate the potential of machine learning approaches to enhance the longevity and reliability of …
-
Scalable Data Paradigms for Steering General-Purpose Language Models
Pretrained Language Models (LMs) have demonstrated remarkable general-purpose capabilities by encoding vast amounts of knowledge from the internet. However, effectively steering these models to serve diverse downstream applications, such as following instructions, chatting with users, using tools, …
-
Artificial neural networks and the cross-section of equity returns: identifying nonlinear opportunities on the Johannesburg Stock Exchange
… architecture is explored, considering varying model depths and node counts. The activation function, training algorithm, learning rate, number of epochs, batch size, and loss function are kept constant across architectures. The findings suggest that portfolios constructed from ANN forecasts …
-
On the Resource Efficiency of Language Models
Large language models (LLMs) have revolutionized a wide range of natural language processing tasks. However, their practical utility faces resource challenges in two dimensions: data efficiency and model efficiency. For post-training, LLMs face data curation challenges where high-quality labeled …
-
Federated Learning With Generalization To New Domains
… that focuses on training machine learning models in a decentralized fashion without having the need to store all data on one central server. In this thesis, we address the challenges of data heterogeneity and label scarcity in FL by proposing two novel approaches for federated domain …
-
A geospatial deep learning framework for scalable hydrographic mapping
… mapping methods based on Digital Elevation Models (DEMs) are constrained by data quality, subjective parameterization, and limited scalability. Recent advances in deep learning and geospatial artificial intelligence (AI) present new opportunities to automate hydrography extraction by …
-
Influence of training dataset selection on the performance of a machine learning model
… located at the University of Saskatchewan. The model has been developed using Deep Learning (DL) based Multi-column Convolutional Neural Network (MCNN) algorithm and TensorFlow framework. This is an object counting model, that counts the Canola flowers from the images based on the learning from …
-
MITIGATING DATA SCARCITY CHALLENGES IN MEDICAL IMAGING ANALYSIS:ADVANCED LEARNING APPROACHES WITH EMPHASIS ON HEMOPHILIC ULTRASOUND IMAGES
… class imbalance, and the adaptation of trained models to different domains (such as knee to elbow transfer). To address the problem of the limited number of total samples, this research investigates the adoption of transfer learning and proposes a new multi-task model to effectively utilize …
-
SMALL AND LARGE PERCEPTION MODELS FOR ROBOTIC NAVIGATION
… strengths of small-scale specialized models, and large-scale vision-language models (LVLMs) with better generalized zero-shot capabilities. Specifically, small-scale models typically focus on specialized tasks, such as object detection, segmentation, and terrain classification. On the …
-
Physics-informed Deep Learning and Differentiable Mechanistic Models for Multicomponent Transport Phenomena
… differential equation (PDE)-based continuum models that introduce a multitude of physical assumptions to improve computational tractability. These assumptions typically oversimplify the governing dynamics posing significant challenges to model generalization at new operating conditions. …
-
Advances in Sparse and Low Rank Matrix Optimization for Machine Learning Applications
… setting, sparse solutions exhibit superior model generalization and have a natural interpretation as conducting feature extraction in high-dimensional datasets. On the other hand, since the rank of a matrix is equivalent to the cardinality of the matrix's vector of singular values, rank can …
-
Essays on Model Selection Uncertainty and Model Averaging: Computational and Empirical Work with Beta Regression, Multiple Linear Regression with ARMA Innovations, and the Minimum Description Length Principle
Uncertainty in model selection is under-explored and frequently resolved non-rigorously through beliefs about generalizability, practical usefulness, and computational ease. This is problematic as model selection routinely admits multiple models which imposes extra uncertainty on all post-selection …
Page 1 of 2