The University of Manchester is overhauling how it trains computer science PhDs, embedding creativity, teamwork, and real-world problem-solving into the traditional deep technical doctorate. Why this matters: The standard UK computer science PhD produces experts in narrow specialisms, but graduates often struggle to communicate across disciplines, work with non-academic users, or think about impact from the start. This gap limits how quickly research findings translate into practical tools—whether in data analysis, energy-efficient computing, or secure smart devices. Potential impact: If this model succeeds, it will produce a generation of computer scientists who can both advance fundamental theory and collaborate effectively with engineers, clinicians, or policymakers. That could accelerate progress in areas that quietly underpin modern life: managing the explosion of data from sensors and networks, building reliable software for multicore processors, and securing the growing web of interconnected smart devices. The centre commits to making this its primary PhD route, adding 10 extra studentships per year and investing in course materials and staff training to sustain the model beyond the grant.
View original technical description
We propose a new model of PhD training which preserves the deep technical research study associated with the UK PhD, while augmenting this with training and practical experience in creativity and innovation, scientific evaluation, and experiences working with and communicating with users from outside academia. We will train the students to think about research impacts from the start of their research. We will give them experiences working in groups with non-academic users to solve real problems, experiences communicating outside their specific research discipline. We will give them opportunities to work in other academic and non-academic research labs. Our goal is to produce PhD graduates who are, on the one hand, skilled experts in their research area and, on the other hand, effective cross-disciplinary communicators and impact-aware, creative innovators.We want to lead in the transformation of the UK computer science PhD, and so will make this our primary PhD. To show our commitment, we will provide an additional 10 studentships per year. We will be able to sustain this model after the grant has finished, because some of the funds will be invested in course and recruitment material development, and staff training.We will use two integrative themes to unify the students and focus them on the challenges facing the next generation of researchers in computer science rather than particular methods associated with a particular research community. These two themes represent broad areas which cut across computer science, and are:Engineering for large and complex data - how to deal with the increasing explosion of data and build models of complex systems. How to use these to extract knowledge, fuse with existing knowledge, and make predictions.Engineering for new technologies - how to optimise the effective and energy-efficient use of multicore processors and interconnected embedded smart devices; explore new uses for smart personalised devices, dealing with security issues and other risks.University of Manchester is an ideal place to run the CDT in Computer Science because:1) Manchester has the UK's broadest computer science research portfolio, from electronic data storage, massively-parallel neural microchip engineering and multicore software through to world-beating theorem provers, machine intelligence, semantic technologies, text-mining and e-Science.2) Manchester has the highest concentration of CS RCUK funding anywhere in the UK.3) Manchester has the longest history in the UK of working with industry, including prototyping world-leading machines for Ferranti and ICL to today's wide portfolio of non-academic collaborators. 4) A strong commitment to this transformation, as evidenced by the large amount of resources we will contribute.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know