Identifying Patterns of Reasoning in Large Language Models

About the Project

There have been significant recent advances in the effectiveness of machine intelligence systems based upon neural networks, deep learning, and machine learning techniques. However questions have been raised about the nature of the reasoning occurring in these systems, how opaque that reasoning is, and how that impacts people's ability to trust the resulting systems. However patterns of human reasoning has been well studied over the last two millennia and a particularly useful approach has been that of argumentation schemes which capture stereotypical patterns of presumptive reasoning as used by people in the form of defeasible arguments.

This project will involve a detailed study of both existing machine intelligence systems and of argumentation schemes. The novel contribution will be an executable model of argumentation schemes and the development of a correspondence theory relating argumentation schemes to machine reasoning. This theory will be used to demonstrate how human and machine reasoning patterns are related, and how the symbolic and numeric approaches inherent to each can be augmented to facilitate the construction of explainable, justifiable, and composable AI systems.

Prospective applicants are encouraged to contact the Supervisor before submitting their applications. Applications should make it clear the project you are applying for and the name of the supervisor(s).

Academic qualifications

A first degree (at least a 2.2) ideally in Computer Science with a good fundamental knowledge of Machine Learning, Argumentation Theory and Artificial Intelligence.

English language requirement

IELTS score must be at least 6.5 (with not less than 6.0 in each of the four components). Other, equivalent qualifications will be accepted. Full details of the University’s policy are available online.

Essential attributes:

• Experience of fundamental Machine Learning techniques
• Competence in ML toolkits, e.g. Scikit-learn, Tensorflow, PyTorch, etc.
• Knowledge of Argumentation Theory
• Good written and oral communication skills
• Strong motivation, with evidence of independent research skills relevant to the project
• Good time management

APPLICATION CHECKLIST

• Completed application form
• CV
• 2 academic references, using the Postgraduate Educational Reference Form (download)
• Research project outline of 2 pages (list of references excluded). The outline may provide details about:

1. Background and motivation of the project. The motivation, explaining the importance of the project, should be supported also by relevant literature. You can also discuss the applications you expect for the project results.
2. Research questions or objectives.
3. Methodology: types of data to be used, approach to data collection, and data analysis methods.
4. List of references.

The outline must be created solely by the applicant. Supervisors can only offer general discussions about the project idea without providing any additional support.

Statement no longer than 1 page describing your motivations and fit with the project.

Evidence of proficiency in English (if appropriate)

To be considered, the application must use

• the advertised title as project title

For informal enquiries about this PhD project, please contact Dr Simon Wells - s.wells@napier.ac.uk