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 11 of 11 for “"Model Serving"”.
-
A large-scale model serving framework
Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2026-05-01
-
Using AI to Improve Price Transparency in Real Estate Valuation
… attributes to enhance traditional Hedonic models. By incorporating Vision Language Models (VLMs) and generative AI, the research evaluates the potential of these technologies to assess non-standard variables like aesthetic appeal, condition and cohesiveness of interior and exterior property …
-
Exploring the cultural understanding of mental illness: A study in the Ovaherero community, Opuwo urban area, Namibia
… analysis, with Arthur Kleinman’s explanatory model serving as the theoretical lens. The findings revealed that mental illness is perceived as an affliction interwoven with spiritual, moral, and social dimensions, often attributed to causes such as witchcraft, ancestral displeasure, and moral …
-
Enabling Efficient ML Inference in SigmaOS with Model-Aware Scheduling
… latency and resource contention, especially as models grow in size and complexity. This thesis addresses the cold start overhead and scheduling inefficiencies of multi-tenant ML serving by integrating the RayServe distributed model-serving framework into σOS, a cloud operating system that …
-
Hosting LLMs on Shared GPUs
Large language models (LLMs) have emerged as powerful tools for a wide array of applications. Serving multiple LLMs on shared GPUs has increasingly gained attention as single providers need to support multiple applications (summarization, chat, code generation), different model versions (A/B …
-
Resource Sharing for Machine Learning Serving
… strides in improving machine learning serving efficiency through various approaches, the rapid evolution of machine learning continues to pose new challenges. Traditional design principles, which focus on optimizing individual components such as computation kernels, memory usage, and …
-
Rethinking Serverless for Machine Learning Inference
… resource utilization. Specifically, we identify model loading and model memory duplication as major bottlenecks in Serverless Inference, and to address these problems, we propose a new approach that rethinks the way we serve FaaS requests. To support this design, we employ a hybrid scaling …
-
Identifying factors associated with bicycle helmet use behavior among college students
… was developed based on the Health Belief Model and the Theory of Reasoned Action, with the Stages of Change Model serving as a tool to classify bicycle riders into groups in accordance with current bicycle helmet use behaviors and future intentions to use. A total of 315 responses were …
-
The structure and development of human-computer interfaces
… a structural, descriptive, language-oriented model of human-computer interaction, based on a theory of human-computer interaction. This model is a design and implementation model, serving as the framework for a dialogue engineering methodology for human-computer interface design and …
-
Towards efficient and reliable infrastructure for machine learning
… particularly large language and generative models, has enabled transformative applications but at immense computational cost. Training and serving these models requires tens of thousands of specialized accelerators that consume megawatts of power and costs hundreds of millions of dollars. …
-
Does the inclusion of climate variables improve tourism demand forecasting performance?
… in econometric and combination forecasting models can improve tourism demand forecasting performance. Climate conditions are important tourism resources which can influence tourists’ decision as to when and where to travel, however, our understanding of the value of climate variables in …