Fully-Funded Global PhD International Scholarship in Computer Science at University of St Andrews

Страна: Великобритания;

Дедлайн: 30.06.2019

Веб-сайт: st-andrews.ac.uk

 

A Global St Andrews Scholarship from St Leonard’s Postgraduate College for students admitted into a jointly-supervised (co-tutelle) programme between the School of Computer Science at the University of St Andrews and the Department of Computer Science at the University of Victoria, Canada. This scholarship offers fully-funded award in the School of Computer Science of the University of St Andrews and funded scholarship at the Department of Computer Science at the University of Victoria in Canada.

The University of St Andrews is pleased to offer a scholarship funded by St Leonard’s Postgraduate College, to support an exceptional student undertaking doctoral research in the following project:

Intelligible and Trustworthy Deep Learning Research for the Humanities

Project description

Machine learning and deep learning techniques offer great promise to extend the reach and comprehensiveness of research in many areas; for example, in research scenarios that focus on the examination of large document corpuses for subtle underlying themes and links. Deep learning can help classify documents according to the scholars’ needs and find document clusters based on complex relationships and multidimensional features.

However, machine learning / deep learning techniques are generally considered “black-box”. This hampers interpretability and critical interrogation of results by non-machine-learning researchers. Without transparency and adjustability of the processes that the machine is carrying out results produced by algorithms cannot be interpreted or trusted. 

This doctoral project aims to address the impediments outlined above through the use of visualization technology as a lingua franca to bridge the gap between machine learning / deep learning algorithm implementation and the needs of the scholars in other areas. Visualization has long been hailed as a fundamental tool to make large amounts of data accessible to humans: visual representations of data allow people to find regularities and insights in an otherwise inaccessible sea of data due to the computational power of the human visual cortex. Visualisation has also been discussed as a mediator between disciplines. 

The general aim of this project is to make machine learning / deep learning technology accessible to researchers in the humanities in a dynamic way that allows for dialog and critical interpretation. The project addresses six driving questions:

  1.  How can deep learning techniques be applied to address the issues of transparency, interpretability, explainability and provenance?
  2. In this context, can visualization serve as a mediating language between scholars in the other areas and deep learning algorithms?
  3. What kinds of deep learning algorithms and workflows can support dynamic, critical and informed interventions by scholars through visualizations.
  4. To what extent this novel integration of visualisation and machine learning can be used by the visualisation designers, data scientists, and even the general public for their own purpose?
  5. What implications has the use of these technologies on the practice of scholars?

 This project has a strong interdisciplinary element, across visualisation and machine learning.

The successful candidate will be supervised by Dr Juan Ye and Dr Uta Hinrichs in the School of Computer Science at the University of St Andrews and by Dr Miguel Nacenta in the Department of Computer Science at the University of Victoria, Victoria, British Columbia, Canada.

Informal enquiries regarding this scholarship may be addressed to Dr Juan Ye – email jy31@st-andrews.ac.uk

 

Похожие гранты и стипендии с открытой подачей заявок: