Foundation Year information
Academic mentors attend the sessions to gauge their mentee’s knowledge and identify areas where further study will be beneficial as part of the Research Orientation module, which completes a student’s foundational training in an individual-centred manner and guides them on the way to choosing a research topic for the Summer Project and subsequent PhD research.
Throughout their programme the students participate in and build an online portfolio of transferable skills training inspired by the Vitae Researcher Development Framework. This includes training in responsible innovation, entrepreneurship, hosting visitors, organising events, public engagement, citizen science, etc. A ‘serious game’ approach will encourage students to build their Researcher Development portfolio, unlocking additional research and training support resources when they reach certain milestones.
Foundations of Practice-Oriented AI
This unit covers the main topics in data-driven AI, knowledge-intensive AI and human-AI interaction. It situates those topics within human-centred and participatory methodologies to bring those AI techniques into practice. Students also study the legal and ethical context ensuring that they know how to deploy AI solutions responsibly and in a manner beneficial to society. As such it forms the foundation of the students’ AI research and practice throughout the programme.
Content will include, but is not limited to, the following topics and discussion questions:
• Responsible AI foundations. Are full transparency and accountability achievable?
• Knowledge representation and reasoning. Can reasoning be emergent?
• Human-in-the-loop AI. Is autonomous AI ever a good idea?
• Large Language Models. Are there fundamental limits to their capabilities?
• Bias and fairness in AI. Is inductive bias always a bad thing?
• Legal and regulatory environment for AI. Does privacy still exist in the age of AI?
• Open data and software. Are companies obstacles for open research?
The Foundations of Practice-Oriented AI module is run in a student-led, ‘flipped classroom’ manner and covers the main topics in data-driven AI, knowledge-intensive AI and human-AI interaction; human-centred and participatory methodologies to bring those AI techniques into practice; and the legal and ethical context that will help students to do so responsibly and in a manner beneficial to society.
Practice Projects in AI
This unit will allow students to engage with application domains provided by CDT partners, directly applying tools and proficiencies they have learned in the programme. Each time the student participates in the unit, they will work in cross-cohort groups, applying an AI solution to a problem originating in a different field. The objective is to learn, through practice, the specific skills needed to conceptualise, design, carry-out, and deploy a particular AI-supported task. Students will disseminate the findings both in written and oral form.
This is mostly a supervised group project, with some training sessions occurring alongside it. The problem is provided and introduced by one of the CDT partners.The supervisors will support students to extend and complement their knowledge and skills gained from other units and activities on the programme to solve a particular task.
Practice projects will be run in close collaboration with CDT partners (academic and industry) and across CDT cohorts to offer ‘deep dives’ into a research domain. Students will liaise with domain experts to get an understanding of their expertise, expectations and success criteria. Exploring different domains will allow the students to acquire the transferable skills that will be key when moving between domains in their future careers.
Summer Project
The main purpose of this initial phase is to compile the literature review and analyse the feasibility, social impact and any ethical issues. . It will deliver a small proof-of-principle implementation and also a report for formal assessment, which includes the outline plan of the subsequent PhD research.