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 385 for “"Generalisation"”.
-
Out-of-distribution generalisation in machine learning
… data, a better grasp of out-of-distribution generalisation is needed for algorithms to be widely deployed and trusted. My thesis presents three research studies that aim to investigate and develop the field of out-of-distribution generalisation. The central goal of these research efforts is …
-
Inductive Biases and Generalisation in Models for Natural Language Processing
… trained on, and are failing to infer correct generalisation strategies from smaller quantities of data. This leads us to ask whether imbuing model architectures with carefully considered inductive biases might lead to improved data-efficiency and performance, by nudging them towards solutions …
-
Studies on the generalisation of Gaussian processes and Bayesian neural networks
… have come to the fore and in this thesis their generalisation capabilities are analysed from theoretical and empirical perspectives. Upper and lower bounds on the learning curve of Gaussian processes are investigated in order to estimate the amount of data required to guarantee a certain level …
-
Triplet entropy loss: improving the generalisation of short speech language identification systems
… investigates several methods to improve the generalisation of language identification systems to new speakers and to new domains. These methods involve Spectral augmentation, where spectrograms are masked in the frequency or time bands during training and CNN architectures that are …
-
Ancestral structures in population genetics and a generalisation to birth-death processes with catastrophes
Ancestral structures have become of paramount importance in the study of population genetics, the branch of biology that investigates the evolution of the genetic composition of populations. In the framework of the two-type Moran model with mutation and selection, we use the pruned-lookdown …
-
Job Sequencing and Tool Switching Problems with a Generalisation to Non-Identical Parallel Machines
… and its variations. Only few authors consider generalisations of the problem because the level of complexity of extended problems is high. A general approach of the job sequencing and tool switching problem with non-identical parallel machines and sequence-dependent setup times is described in …
-
Unmasking the imposters: towards improving the generalisation of deep learning methods for face presentation attack detection.
… to enhance the performance of detection and generalisation further. This problem is motivated by the need to achieve generalization against new and rapidly evolving attack variants. It is possible to extract identifiable features from presentation attack variants in order to detect them. …
-
Is implicational generalisation unidirectional and applicable in foreign contexts? : evidence from relativization instruction in a foreign language
… relativization and examined the Markedness Generalisation Hypothesis (MGH) that was generated from the studies mentioned above. The first research question addressed was: Do the students who get instructed on relative clauses perform better than those who do not get any of that instruction? …
-
Symbolic and semantic fear and avoidance generalisation in humans: An examination of boundary conditions and convergence with trait measures
… conditioning and the symbolic or semantic generalisation of fear and avoidance. A small number of previous studies had already attempted to correlate empirically observed levels of generalised threat and avoidance responding with scores on a number of trait and experiential avoidance …
-
Recombinative Generalisation and Psycholinguistic Grain Size: An Investigation of Reading Related Skills in Adult Readers and Children with Learning Difficulties
… novel patterns is referred to as recombinative generalisation. In the case of reading, sub-word units (e.g., onsets and rimes, phonemes, syllables) of known words are recombined to form new words that can be pronounced by reference to letter–sound correspondences. Matching-to-sample (MTS) …
-
Learning, Consolidating, and Generalising Novel Morphology
… morphological knowledge in supporting linguistic generalisation, the neural representations supporting its learning remain largely unexplored. This thesis addressed this gap by exploring the role of memory consolidation in morphological learning and generalisation. In three experiments, adult …
-
Bi-fractional transforms in phase space
… and Weyl functions (displacement operator). The generalisation of the parity-displacement operator relationship considered here is called the bi-fractional displacement operator, O(α, β; θα, θβ). Additionally, the bi-fractional displacement operators lead to the novel concept of bi-fractional …
-
Efficient Adaptation of Large Language Models in Natural Language Processing
… Information Retrieval (IR). Despite their strong generalisation capabilities, LLMs still require domain- and task-specific fine-tuning to achieve competitive performance in highly specialised scenarios. LLMs exhibit some generalisation abilities, though these are not uniformly reliable across …
-
The structure and stability of ecological networks
… level structures. Chapter 2 deals with whether generalisation drives abundance or vice versa in plant-hummingbird pollination networks. I find evidence that abundance drives generalisation and use a simple model to show that neutral processes can explain broad patterns of species- level …
-
Generalised Bayesian matrix factorisation models
… vision, amongst others. This thesis develops generalisations of matrix factorisation models that advance our understanding and enhance the applicability of this important class of models. The generalisation of models for matrix factorisation focuses on three concerns: widening the …
-
Modelling non-linearity in 3D shapes: A comparative study of Gaussian process morphable models and variational autoencoders for 3D shape data
… (VAE). Their model performance is measured using generalisation, specificity and computational efficiency in training. The research showed that, given limited computational power, GPMMs managed to achieve improved relative generalisation performance compared to VAEs, in the presence of non-linear …
-
Continuous Representations in Machine Learning - With applications to medical imaging and operator learning
… theoretical understanding and interpretability, generalisation capabilities particularly in low data regimes, robustness to noise and training in restricted resource settings. Many real-world observations can naturally be described as continuous processes rather than discrete data points -- …
-
The causal role of selective information processing biases towards threat in anxiety
… and the emotional Stroop task (measure of generalisation of training). The 54 undergraduate student participants, who were in a mid-range of trait anxiety, were randomly allocated to one of 3 orders in which to receive these tasks pre and post training. There was a failure to train …
-
A comparative study of fashion design processes in the U.K. women's wear industry (Volume I)
… between the variables and culminated in a generalisation about the variables within the fashion design processes and their market segmentation.
-
Generalised, multilingual, optical Braille recognition models
… in training data leads to improved performance generalisation, for both modelling approaches. In particular, a multi-objective resampling strategy minimising both the imbalance in character classes and the correlations between Braille dots, yields exceptionally stable performance across …
Page 1 of 20