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 148 for “"Model Reliability"”.
-
Improvement of model reliability in a soil-plant system
本文データは平成22年度国立国会図書館の学位論文(博士)のデジタル化実施により作成された画像ファイルを基にpdf変換したものである
-
ENHANCING SEQUENCE MODEL RELIABILITY: STUDIES ON NOISE ROBUSTNESS, DEMOGRAPHIC FAIRNESS, AND ATTENTION SINK
Modern Transformer-based sequence models perform well in controlled settings but degrade under real-world conditions such as noise, demographic variability, and long contexts. This thesis studies model reliability across singing voice transcription (SVT) and language models (LMs). First, we address …
-
Integrated System Model Reliability Evaluation and Prediction for Electrical Power Systems: Graph Trace Analysis Based Solutions
A new approach to the evaluation of the reliability of electrical systems is presented. In this approach a Graph Trace Analysis based approach is applied to integrated system models and reliability analysis. The analysis zones are extended from the traditional power system functional zones. The …
-
Towards Post-Hoc Human-Interpretability of Multimodal Neural Networks for Healthcare Applications
… the decision-making logic of these large AI models is often unknown. This raises serious concerns surrounding model reliability, accountability, patient autonomy, and bias. The black-box nature of these models often makes them unsuitable for high-risk healthcare applications. Research in …
-
Towards Safe And Reliable AI Models
… in our daily lives. Emerging advanced AI models, from general-purpose Large Language Models (LLMs) such as ChatGPT and Deepseek to domain-specific systems such as autopilot for autonomous vehicles, are changing the way we work and live by increasing efficiency and fostering innovation. …
-
Towards efficient and reliable neural networks
… applications like autonomous driving, deployed models must be reliable. In this thesis, we tackle these challenges with four main contributions:</p> <p>The first contribution addresses the constraints in computational resources via pruning at initialization. Unlike traditional pruning methods …
-
Reliability of Machine Learning Models in the Real World
… settings is limited due to concerns about reliability. In this paper, we’ll first explore methods to verify a model’s reliability in diverse scenarios, including classification, detection, auctions, and watermarking. We’ll then discuss the challenges and limitations of these verification …
-
Reliability and Availability Analysis of Mining Systems with Human-Error
This study presents reliability and availability analysis of mining systems with hardware and human-error failures. The failures of proposed models representing the mining systems could be safe or unsafe. To understand model reliability and availability behaviours, different values of failure and …
-
Improved integration of information to reduce subsurface model bias
Subsurface modeling deals with data-related issues like cognitive and sampling biases, and model-related challenges including statistical assumptions, misspecification, and algorithmic biases. These challenges introduce four critical implications during subsurface modeling. Firstly, subsurface …
-
Modeling the population dynamics of herring in the Prince William Sound, Alaska
… and conducted by reformulating the current model used by Alaska Department of Fish and Game to manage this population. The pre-season spawning biomass of herring age-3 and older in 2013 was estimated to be between 12,150 and 31,740 mt (95% credibility interval) with a median of 19,410 mt. …
-
Seismic traveltime inversion of wide-angle data for strongly-varying structure: Central Chilean margin and the subducting Juan Fernandez Ridge
… The traveltime inversion method was used to model the data and assess the model reliability. A tomographic approach was used to assess the objectivity of the structures in the final models. The final models include slope sediments, the Valparaiso forearc basin, subducting sediments, an …
-
Communication protocols, queuing and scheduling delay analysis in CANDU SCWR hydrogen co-generation model
… load. CANDU-SCWR hydrogen co-generation model reliability can be analyzed by dynamic flow graph methodology. We have analyzed the CANDU-SCWR feed water integration with the oxygen unit of copper chloride cycle and also conducted an analytical review of the current networked control system …
-
Robustness of ai-driven histopathology under real-world adversarial examples
… robustness of histopathology classification models under adversarial and real-world perturbations resembling clinical artifacts is being investigated.</p><p>Using whole-slide images from the CAMELYON17 cohort, four representative architectures—ResNet-18, ResNet-50, HIPT-2MLP, and ViT-B/16 …
-
Constitutive Property Testing and Reliability Assessment of Lead-Free Solder Joint
… parameters are obtained for (1) a partitioned model, where the elastic, rate-independent plastic and rate-dependent creep behaviors are all separated and (2) a unified creep-plastic model. Reliability assessment of lead-free solder joint is performed by calibration of virtual qualification …
-
Seismic site characterization through joint modeling of complementary data functionals, with applications to Santo Domingo, Dominican Republic.
… seismic "site characterization" through joint modeling of horizontal to vertical spectral ratios (HVSR) and surface wave dispersion, performed via refraction microtremor (ReMi). Fitting of data functionals by synthetics is driven by global optimization. The products of this approach are shear …
-
Robot reliability through fuzzy Markov models
… robots have increased the importance of robotic reliability and fault tolerance. Standard approaches of reliability engineering rely on the probability model, which is often inappropriate for this task due to a lack of sufficient probabilistic information during the design phase. Fuzzy logic …
-
Towards Reliable Image Classification: A Systematic Robustness Analysis of CNN and Classical Models Under Natural Corruptions
Machine learning models, particularly Convolutional Neural Networks (CNNs), dominate image classification tasks in critical domains such as medical imaging, autonomous driving, and insurance. However, despite high accuracy on clean benchmark datasets, these models often exhibit significant …
-
An Investigation of University General Education Assessment Practice
Many-facets Rasch modeling (MFRM) is an effective way to compensate for rater effects (e.g., rater leniency/severity) in the ratings assigned during rater-mediated assessment. While a fully crossed MFRM rating design, in which all raters rate every rating object, is the ideal from a psychometric …
-
Modelling non-linearity in 3D shapes: A comparative study of Gaussian process morphable models and variational autoencoders for 3D shape data
… variation in 3D data is known to influence the reliability of linear statistical shape models (SSM). This problem is regularly acknowledged, but disregarded, as it is assumed that linear models are able to adequately approximate such non-linearities. Model reliability is crucial for medical …
-
Reliability Analysis of Carbon Fiber Reinforced Polymer (CFRP) Strengthened Steel Beams
… techniques used to produce the joints. Reliability analysis of adhesively bonded CFRP-to-steel double-lap shear (DLS) joints with thin outer adherends was performed first, to study the bond behavior when dominated by adhesive shear stresses. A comprehensive experimental database of the …
Page 1 of 8