Funding providers: Engineering and Physical Sciences Research Council (EPSRC) DTP and Swansea University's Faculty of Science and Engineering
Subject areas: Computer Science / Artificial Intelligence
Project start date:
- 1 October 2023 (Enrolment open from mid-September)
- Dr Joseph MacInnes
- Professor Xianghua Xie
Aligned programme of study: PhD in Computer Science
Mode of study: Full-time
The addition of 'attention' to machine learning has recently improved many algorithms with the potential to transform a wide range of machine learning approaches (transformers, perceivers). Attention, simply, allows an algorithm to allocate more weight to input that is relevant for certain tasks, and less weight to the irrelevant. Attention mechanisms in machine learning will become increasingly important as the volume of input data increases, and even efficient algorithms will have to make informed choices about which input should receive priority or actively inhibited. We will apply current theories of human attention to improve machine learning algorithms that adjust to the goals of the agent. We will use high quality eye tracking data in various tasks as a proxy for human attention and use these data to inform novel attentional mechanisms for machine learning.
Candidates must normally hold an undergraduate degree at 2.1 level in Computer Science, Mathematics or a closely related discipline, or an appropriate master’s degree with a minimum overall grade at ‘Merit’ (or Non-UK equivalent as defined by Swansea University).
English Language requirements: If applicable – IELTS 6.5 overall (with at least 6.0 in each individual component) or Swansea recognised equivalent.
This scholarship is open to candidates of any nationality.