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 “"Computer Science (miscellaneous)"”.
-
A Multi-Scale Transformer Architecture for Cardano Price Forecasting with Retrieval-Augmented Context
Forecasting cryptocurrency prices has been a challenge due to their high volatility, non-stationary and regime-dependent dynamics. Recent advances in deep learning, particular Transformer architectures, have shown their superiority in time series forecasting by leveraging self-attention to capture …
-
Data collection through design optimization methods in MABTs-based experiments
periments to study human decision-making. Due to the increased computational power available to researchers and the possibility of implementing online experiments with mobile applications, there is a wide selection of experimental designs for these tasks. Design optimization (DO) methods can help …
-
Blockchain-Enabled Reinforcement Learning for Privacy-Preserving and Trust Management in Vehicular Networks
Vehicular networks have become a fundamental component of next-generation intelligent transportation systems, with the Internet of Vehicles (IoV) enabling communication among vehicles and infrastructure. While this connectivity offers significant benefits, the large scale of participating vehicles, …
-
Enhancing Microsleep Prediction Performance Through Temporal Modeling, Multimodal Fusion, and Label-Efficient Learning
… biomedical signal analysis, and related brain–computer interface applications.
-
Towards Modular Controlled Text Generation: Attribute-Specialised PEFT Modules and Comparative Evaluation
Controlled Text Generation (CTG) aims to produce text that reliably adheres to specified attributes, such as sentiment, topic, or factual content. Despite the impressive generation capabilities of large language models (LLMs), standardised evaluation protocols are largely absent, simultaneous …
-
A Neuro-Symbolic Framework to Enhance Interpretability of Graph Neural Networks
Graph Neural Networks (GNNs) are now a standard tool for modelling graphstructured data in applications such as molecular property prediction, drug discovery, recommender systems, and citation networks. In high-stakes domains, especially computational biology and chemoinformatics, practitioners …
-
Enhancing Lifelog Question-Answering with Retrieval-Augmented Generation and Reasoning Large Language Models
Lifelogging refers to the process of passively collecting, storing, and analysing an individual’s daily life data from wearable cameras, sensors, and mobile devices. Question-answering (QA) on lifelog data involves posing questions on lifelog data and generating answers in natural language. Lifelog …
-
Modularity for Reusable Large Language Models: External PEFT Modules and Internal Circuits
Large language models (LLMs) are powerful, but costly to train, adapt, and deploy when treated as monolithic systems. This thesis studies reusability and efficiency under explicit constraints on parameters, floating-point operations (FLOPs), memory, data, and deployment cost. It treats modularity …
-
Machine learning approaches to uncover the neural mechanisms of motivated behaviour: from ADHD to individual differences in effort and reward sensitivity
Motivated behaviour relies on the brain’s capacity to evaluate effort and re- ward. Dysregulation within these processes contributes to a spectrum of conditions, from hyperactivity in attention-deficit/hyperactivity disorder (ADHD) to dimin- ished goal-directed behaviour in apathy – characterised …
-
Towards Improved Conversation Context Understanding for Effective Conversational Information Retrieval
Conversational Information Retrieval (CIR) has garnered considerable attention in recent years. A key challenge in CIR is formulating effective queries during the dialogue between the searcher and the conversational agent. Unlike traditional information retrieval systems, where users independently …
-
Designing For Cognitive Resilience: Sustaining Attention, Memory, And Reasoning In Everyday Digital Interaction
Everyday digital technologies are designed to maximise speed, ease of use, and continuous engagement. While these qualities support traditional usability goals, they can also undermine the cognitive capacities they were designed to support. This thesis introduces cognitive resilience as a …