Sort by
Refine Your Search
-
Listed
-
Category
-
Program
-
Field
- Computer Science
- Medical Sciences
- Economics
- Engineering
- Business
- Law
- Science
- Biology
- Education
- Materials Science
- Humanities
- Linguistics
- Mathematics
- Arts and Literature
- Philosophy
- Environment
- Sports and Recreation
- Chemistry
- Design
- Earth Sciences
- Electrical Engineering
- Psychology
- Social Sciences
- 13 more »
- « less
-
Background and Motivation Modern deep learning models have achieved remarkable success in computer vision and natural language processing. However, they typically produce overconfident predictions
-
Species’ distributions are shifting in response to global climate change and other human pressures. Accurate methods to monitor and predict distribution shifts are urgently needed to manage
-
aims to explore who takes physics and astrophysics major units, why they pursue them, and what obstacles they may face. There are a number of research questions under this umbrella. Computational
-
events with the GOTO telescope network. Projects focussing on thermonuclear bursts will involve analysis of new and archival data from satellite-based X-ray telescopes, and running numerical models
-
analysis, contextual analysis, audio feature extraction, and machine learning models to identify and assess potentially dangerous content. Similarly, computer vision models are implemented to analyse images
-
tissues or reveal micro- or nano-structural features, like the small air sacs in lungs. To overcome these limitations, alternative X-ray imaging methods have been developed: X-ray phase-contrast and dark
-
the optical-to-radio wavelength range, from major surveys and space telescopes (e.g: Gaia, SDSS, JWST, Hubble, Roman, Rubin-LSST). These are analysed using advanced machine learning and data-driven methods. My
-
methods dealing with model complexity - e.g., AIC, BIC, MDL, MML - can enhance deep learning. References: D. L. Dowe (2008a), "Foreword re C. S. Wallace", Computer Journal, Vol. 51, No. 5 (Sept. 2008
-
. Recent works on knowledge graph question generation [4,5] have mainly focussed on multi-hop questions. This project aims at developing novel methods that jointly address the challenging, dual problem
-
privacy constraints, robust solutions are essential. This PhD project will develop methods for building reliable medical imaging models that generalize across distribution shifts without retraining