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 51 for “"Robust Generalization"”.
-
InfoMixup : An Intuitive and Information-Driven Approach to Robust Generalization
… InfoMixup , a novel method to improve the robustness of Deep Neural Networks without significantly affecting performance on clean samples. Our work is focused in the domain of image classification, a popular target in contemporary literature due to the proliferation of Deep Neural Networks …
-
Toward a Unified Framework for Open World Visual Learning
… that enables visual learning systems to operate robustly in dynamic and unpredictable environments. The research addresses three fundamental challenges: unknown awareness, efficient adaptation, and robust generalization. First, to improve unknown awareness, we develop a weakly supervised …
-
Adversarial Robustness of Deep Learning Models
… training (AT) and its variants aim to enhance robustness by incorporating adversarial examples into training. However, AT often leads to both standard and robust generalization issues, the causes of which remain largely elusive due to the complex learning dynamics involved. This thesis …
-
Toward the Generalization of Vision Models
… ability to generalize is essential for ensuring robust and reliable performance across diverse datasets and real-world scenarios. Generalization refers to a computer vision model's capability to accurately interpret and process new, unseen data beyond the specific examples it was trained on. …
-
Automating computed tomography diagnostic image quality control using deep learning
… state-of-the-art deep learning models capable of robust generalization on medical imaging classification tasks, and 3) the absence of a dedicated deep learning backbone pre-trained on medical imaging data capable of robust generalization for various medical image analysis tasks. This thesis first …
-
Ranking the Cosmos : identifying strongly lensed galaxies
… Integrating real and simulated data ensures robust generalization to real observations. This study demonstrates that simpler models like SVMs can outperform complex deep learning architectures for this task, offering a promising alternative for classifying strongly lensed galaxies in large …
-
Learning-based MRI Reconstruction Method with Coil Sensitivity Estimation and Prior Adaptation
… methods in terms of PSNR and SSIM, showcasing robust generalization to varying acceleration rates and imaging modalities. Our approach thus provides an effective and flexible solution for accelerated MRI reconstruction in clinical scenarios.
-
AdFair-CLIP: Adversarial Fair Contrastive Language-Image Pre-Training for Chest X-Rays
… and diagnostic accuracy, while maintaining robust generalization in zero-shot and few-shot scenarios. These results establish new benchmarks for fairness-aware learning in CLIP-based medical diagnostic models, particularly for CXR analysis.
-
Structure-Based Prediction and Design of Adaptive Immune Receptors Targeting Peptide–MHC Complexes
… and structure prediction frameworks enable both robust generalization and precise molecular engineering. These tools lay the foundation for programmable, therapeutic recognition of diseased cells and expand our understanding of adaptive immune specificity.
-
UNSUPERVISED DEEP ANOMALY DETECTION AND ITS APPLICATION
… models, ensuring broader applicability and robust generalization.
-
A GAN-Augmented Machine Learning Framework for Predicting Raman Characteristics in Carbon Nanofiber Synthesis
… % marginal fidelity / relationship), allowing robust generalization despite the small-n regime.
-
Credit rating prediction using multi-modal temporal neural networks
… both seen and unseen data scenarios demonstrates robust generalization capabilities, essential for practical deployment in financial risk assessment.
-
Addressing Data Scarcity with 2D Projection-Based 3D Point Cloud Semantic Segmentation
… a large volume of training data to achieve robust generalization and prevent overfitting. However, acquiring extensive 3D datasets remains a significant challenge. This study addresses data scarcity in 3D semantic segmentation by leveraging a 2D projection-based approach, utilizing …
-
COMPOSITIONAL GENERALIZATION IN INSTRUCTION FOLLOWING TASKS
… This dissertation studies the compositional generalization abilities of machine learning models in various instruction following tasks. We study the various dimensions of compositionality for a diverse set of instruction follow- ing tasks of varying complexity: semantic parsing in synthetic …
-
Mining Attributed DTE Policies
… We discovered a metric that produces good generalization performance and supports modular training. Policy mining is then implemented by a reduction to MaxSAT. We devised an efficient MaxSAT encoding known as the separator-based encoding. Using both real-world and synthetic datasets, we …
-
Advancing Wireless Sensing in Granularity, Efficiency, and Generalization Across Diverse Wireless Technologies
… in sensing and communication integration, and robust generalization across diverse domains. This thesis addresses these challenges through three contributions spanning mmWave radar, 5G, and Wi-Fi. First, to enable fine-grained interaction, mmFinT presents a 3D hand pose tracking system using …
-
METHODOLOGICAL ADVANCES IN REAL-WORLD COMPUTER VISION: ROBUSTNESS, EFFICIENCY AND HUMAN PREFERENCE ALIGNMENT
… models into real-world applications: inference robustness, iteration efficiency, and human-aligned evaluation. First of all, given that noise and corruption are prevalent in real-world data, it is important for practical computer vision models to be robust against input perturbations. To this …
-
Curriculum-based soft actor-critic for multi-section roll-to-roll tension control
… exposure enables stable policy learning and robust generalization without requiring explicit system identification or controller retuning for different operating conditions. Simulation studies on a three-section R2R system demonstrate that the curriculum-trained SAC policy achieves …
-
Emergent Syntactic Behaviors and Mechanisms in Neural Language Models
… of phrases and sentences—is necessary for robust generalization in natural language processing tasks. One reason for the success of large pre-trained language models (LMs) is that pre-training teaches LMs to rely on syntactic features, rather than incorrect surface features like word …
-
Curating the world’s largest biodiversity dataset for AI
… classification, taxonomic diversity, and robust generalization to unseen species. This work has received substantial recognition, including a \textbf{Spotlight presentation at NeurIPS 2024} and national-level coverage by the \textbf{USDA National Institute of Food and Agriculture (NIFA)}. …
Page 1 of 3