Integrating structured knowledge with Machine Learning Representations
he problem: Machine learning systems are effective at learning from scratch, but it is not clear how to reuse the knowledge they have learned for another purpose. The knowledge is “entangled” in the network, and tied to the task the system was trained for. In addition, it is also not clear how a typical machine learning system, could take advantage of existing prior conceptual knowledge; instead they usually learn from scratch.
The project: This project will investigate how conceptual knowledge can be integrated with the representations of typical machine learning systems. This is a combination of the symbolic with the subsymbolic, with the aim of achieving a more human-like cognition. We aim to come up with representations which are, at a high level, composed of parts meaningful to humans, but which also connect to subsymbolic representations at a low level, so that they can integrate with typical learning techniques. This project will build a more human-like representation of a set of concepts, grounded in subsymbolic representations. The project will also build a system for conceptual reasoning, which can apply these concepts to do reasoning with real data in language or computer vision. Our representations will build on cognitive science works that provide alternative views of concepts. Although the project is to be based at Aberdeen, we will also collaborate with experts outside Aberdeen.
Candidates should have (or expect to achieve) a UK honours degree at 2.1 or above (or equivalent) in Computer Science or Informatics. Other closely related backgrounds can also be considered, like Engineering or Mathematics with computer algorithms experience (e.g. in Matlab).
It is essential that the successful applicant has a background in Computer Science knowledge (programming, algorithms, data structures, confident in machine learning relevant mathematics. Knowledge of programming is important and some knowledge of machine learning is desirable.
• Apply for Degree of Doctor of Philosophy in Computing Science
• State name of the lead supervisor as the Name of Proposed Supervisor
• State ‘Self-funded’ as Intended Source of Funding
• State the exact project title on the application form
When applying please ensure all required documents are attached:
• All degree certificates and transcripts (Undergraduate AND Postgraduate MSc-officially translated into English where necessary)
• Detailed CV
• Details of 2 academic referees
Informal inquiries can be made to Dr F Guerin ([Email Address Removed]@abdn.ac.uk) with a copy of your curriculum vitae and cover letter. All general enquiries should be directed to the Postgraduate Research School ([Email Address Removed])
This project is advertised in relation to the research areas of the discipline of Computing Science. The successful applicant will be expected to provide the funding for Tuition fees, living expenses and maintenance. Details of the cost of study can be found by visiting View Website. THERE IS NO FUNDING ATTACHED TO THESE PROJECTS.