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 700 for “"Deep learning (Machine learning)"”.
-
Applications of Deep Learning, Machine Learning, and Remote Sensing to Improving Air Quality and Solar Energy Production
… from Planet Labs. In this study, we employ a deep convolutional neural network (CNN) to process the imagery by extracting image features that characterize the day-to-day dynamic changes in the built environment and more importantly the image colors related to aerosol loading, and a random …
-
A Study of Compression Methods and Their Applications to Large Geological Models
The promise of deep learning is to discover rich hierarchical structure of the data that allows to circumvent manual feature extraction and engineering. In this study, we explore novel applications of the deep learning, building on the representational power of the neural networks to find latent …
-
ncRNA-protein Interaction Prediction using Language-based Features
… interaction (RPI). Several classic and deep-learning machine learning models have been pro-posed to predict RPI. These models first need to collect features of RNA and protein, such as physicochemical properties, secondary and tertiary structure, et cetera, before feeding them into the …
-
Advancing Natural Resource Management Through Artificial Intelligence
… artificial intelligence (AI)—including deep learning, machine learning, and neural networks— were evaluated for their potential to streamline data management. Three AI-powered prototypes were developed to automate the conversion of written, audio, and visual data to structured …
-
Neurocube: Energy-efficient programmable digital deep learning accelerator based on processor in memory platform
Deep learning, machine learning algorithm based on artificial neural network, shows great success in numerous pattern recognition problems, such as image recognition or speech recognition. Most of deep learning developments are based on the software platform with general purpose graphic processor …
-
Where are the objects? : weakly supervised methods for counting, localization and segmentation
In 2012, deep learning made a major comeback. Deep learning started breaking records in many machine learning benchmarks, especially those in the field of computer vision. By integrating deep learning, machine learning methods have became more practical for many applications like object counting, …
-
Improving Pedestrian Safety Using Video Data, Surrogate Safety Measures and Deep Learning
… using video data, surrogate safety measures and deep learning. Machine learning (including deep learning) models are proposed for predicting pedestrians' potentially dangerous situations. On the one hand, pedestrians' red-light violations can expose the pedestrians to motorized traffic and pose …
-
Deep Learning Models for Gait Analysis and Treatment Planning for Children With Cerebral Palsy
… treatment plan for a child with cerebral palsy, deep learning machine learning models were developed to quantify gait quality to recommend procedures that can best improve gait quality. 1) A deep learning autoencoder was used to compress the person's instrumented gait data into a single value, …
-
Essays on Examining Financial Markets' Dynamics and Forecasting by Deep Learning and Econometrics Models
… financial crises. For the second topic, several deep learning, machine learning, and hybrid models are adapted to improve the forecasting of crude oil, gold, and silver markets. For this purpose, I implemented sixteen different deep learning and machine learning models on historical price data …
-
3D coronary artery tree reconstruction from two non-simultaneous angiographic projections using deep learning
… This thesis aims to develop automated deep learning-based approaches for reconstructing 3D coronary artery trees from only two clinical ICA projections acquired from a dual-axis rotational single-plane C-arm system, while accounting for realistic acquisition constraints. By reducing …
-
Neural networks as data
<p>In machine learning, neural networks are typically treated as models to be used for various tasks such as classification, regression or generative modeling. In this thesis, we take a different perspective and treat neural networks as data instead of models.</p> <p>Indeed, in deep learning, data …
-
Advancing data-efficient deep learning: non-parametric transformers, active testing, and in-context learning
… an important role in the practical success of deep learning. Unfortunately, in many real-world scenarios, high quality data may be scarce, such that the naive application of deep learning can fall short of expectations. A large variety of prior work aims to remedy this and make deep learning …
-
Generative modelling: addressing open problems in model misspecification and differential privacy
… imputation. After categorising current deep generative imputation approaches into the classes of nonignorable missingness models as introduced by Rubin [1976], we extend the formulation of variational autoencoders to factorise according to a nonignorable missingness model class that has …
-
Towards robust machine learning with graph neural networks
… data, or generative methods that rely purely on learning and often fail to generate adversarial examples where they are hard to find. To alleviate these deficiencies, we propose a novel stand-alone attack based on a GNN that takes advantage of the strengths of both approaches. Our GNN computes …
-
High performance deep learning on resource constrained platforms
… accelerated progress in the different areas of machine learning (ML), it has become virtually impossible to not interact with ML-powered applications and services in our everyday lives. Recommender systems, computational photography, speech translation or, getting directions are only a handful …
-
Understanding Gaussian noise injections in neural networks
<p>Deep Learning (read neural networks) has emerged as one of the most exciting and powerful tools in the statistical machine learning toolbox. Neural networks over the past decade have defeated incumbent state of the art models in computer vision, natural language processing, and reinforcement …
-
Universal in-context approximation
… from task-specific training to in-context learning via prompting. This raises a fundamental question: can a model with fixed weights solve novel tasks as effectively as a fine-tuned one? This thesis investigates the theoretical capabilities and limitations of in-context learning across …
-
Efficient large language models for the NHS and psychiatry
… Efficient adaptation methods, such as prompt learning, were devel- oped and employed, reducing computational and storage requirements by up to 98% while maintaining state-of-the-art performance on several clinical NLP tasks. The bespoke NHS LLMs and efficient adaptation methods were applied to …
-
Scalable continual deep learning with computational cost considerations
<p>Continual Learning (CL) is a emerging field that focuses on developing models capable of learning continuously from a incoming stream of data, as opposed to hundreds of passes over static, curated datasets. These models aim to retain previously acquired knowledge while seamlessly integrating new …
-
Counting with limited supervision
… to be counted is completely unknown. While prior machine learning-based methods have addressed the problem of counting previously unseen kinds of objects, known as class-agnostic counting, they have all required both user input during deployment in the form of exemplar images to define the type to …
Page 1 of 35