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 64 for “"Supervised learning (Machine learning)"”.
-
Grey-Box Bayesian Optimization: Improving Performance by Looking Inside the Black-Box
… hyperparameters for iteratively trained supervised learning machine learning algorithms; and optimizing non-convex functions with a new efficient multistart stochastic gradient descent algorithm.
-
Virtual Multimodal Object Detection and Classification with Deep CNNs
… image modality, when used in tandem as inputs to supervised learning (machine learning) tasks. We begin with a simple score fusion scheme, based on the OpenBR face recognition suite, that evaluates similarity (match-scores) between subject faces in different images. Combining the face matching …
-
On the adversarial robustness of Bayesian machine learning models
Bayesian machine learning (ML) models have long been advocated as an important tool for safe artificial intelligence. Yet, little is known about their vulnerability against adversarial attacks. Such attacks aim to cause undesired model behaviour (e.g. misclassification) by crafting small …
-
A predictive and scalable health-monitoring system for handpumps
… retrospectively assessed on the cloud using two supervised machine learning methods: support vector classifier (SVC) and random forest (RF) and compared to conventional machine learning methods: multilayer perceptron (MLP) and 1-D convolutional neural network (CNN). The novelty filter threshold …
-
Using machine learning to discover candidate localised transcripts from microscopy and genome-wide bioinformatics data
… localisation in glia and additionally I am using supervised machine learning models to predict candidate genes that could potentially be localised in glial cell protrusions.</p> <p>Firstly, I present a YFP protein trap microscopy screen conducted collaboratively by members of the Davis lab for 200 …
-
Knowledge transfer techniques for dynamic environments
… labels for data has led to the emergence of semi-supervised learning techniques which try to make use of both the labeled and the unlabeled data to obtain classifiers with better generalization capabilities. Most existing semi-supervised methods assume that the unlabeled data have the same …
-
Hash kernels and structured learning
… to process massive data remains a challenge for machine learning algorithms. We propose hash kernels to facilitate efficient kernels which can deal with massive multi-class problems. We show a principled way to compute the kernel matrix for data streams and sparse feature spaces. We further …
-
Adversarial robustness of Bayesian neural networks
… In theory, applying Bayesian inference to the learning of neural network parameters carries the promise of solving many practically vexing problems that arise under the frequentist learning paradigm. In particular, Bayesian learning allows for principled architecture comparison and selection, …
-
Fast multi-labelling in early vision
Multi-labelling algorithms are widely used in solving early vision problems. These early vision problems, which include image segmentation, stereo correspondence estimation, etc., are of fundamental importance in computer vision research and applications. Solving a multi-label problem is in general …
-
Learning for the internet : kernel embeddings and optimisation
In this thesis, we develop principled machine learning methods suited for complex real-world Internet challenges. The Internet has supplied an unprecedented amount of data; the challenge now is to transform this massive amount of data into information that supports knowledge creation. Machine …
-
Towards robust neural networks: evaluation and construction
<p>With their supreme performance in dealing with a large amount of data, neural networks have significantly benefited our lives in many aspects, from home assistants to autonomous driving cars. However, it is found that neural networks are brittle. By slightly perturbing the inputs in a way …
-
Quantifying spatial heterogeneity in tumours from dynamic contrast-enhanced MRI
Cancerous tumours are characterized by an inherent perfusion heterogeneity that can change over time and following therapy. The process of angiogenesis often results in increasingly disorganised and leaky vasculature as tumours progress. This may lead to heterogeneous perfusion, which is known to …
-
Annotation efficient learning with affinity graphs
Annotation-efficient learning has emerged as a critical area of research due to the scarcity of labeled samples posing a substantial barrier to developing robust fullysupervised deep neural networks. When the availability of labeled samples is limited, developing an effective learning system …
-
Machine learning based warning system for failed procurement classification documents
Warning systems in the Machine Learning field of study, is a tool that generates a warning based on a model’s prediction results. This thesis’s study topic is to create such system to identify possible problematic procurement classification documents. Given a database of a company, a dataset was …
-
Some contributions to semi-supervised learning
Thesis (Ph. D.)--Michigan State University. Department of Computer Science, 2010
-
Advanced neural network methods for atomistic materials chemistry
Machine learned force fields (MLFFs) are functions trained to map the positions of atoms within a material or molecule to the potential energy and atomic forces generated by expensive, quantum-mechanical calculations. Their high accuracy and low cost are enabling faithful simulations of matter at …
-
Aiding diagnosis of rare diseases from photographs using machine learning
<p>The objective of this thesis is to further research in the field of computational syndrome diagnosis from ordinary photographs of patient faces. As well as providing insight into the phenotypic scope of specific genetic disorders, the development of such tools may eventually help to shorten the …
-
Learning and inference over relational data
… models have enabled significant breakthroughs in learning representations over relational structures. However, the abilities and limitations of these systems are not completely understood, and several challenges remain to endow these models with reliability guarantees, enrich their relational …
-
Predicting fluoroquinolone resistance in Mycobacterium tuberculosis
… of mutations on fluoroquinolone resistance; machine learning algorithms and free energy calculation.</p> <p>I find that fluoroquinolone resistance is widespread and the associations with resistance are complex; genetic and geographic background, resistance to other antitubercular drugs and …
-
Improving neural networks by reduction of parameters and noise injection
<p>Deep learning has been revolutionising the way we look at prediction, signal processing, perception, classification, RL, control and many others. Neural networks have become the centre of attention of the scientific community re- volving around computational solutions to difficult problems.</p> …
Page 1 of 4