The School of Computer Science, here at The University of Sheffield, is set to play a leading role in the UK’s new Innovation and Knowledge Centre (IKC) for neuromorphic computing, following a £12.8 million investment from UK Research and Innovation (UKRI).
Set to launch in October 2025, the UCL-led centre, Neuroware, will accelerate the development of brain-inspired semiconductor technologies that promise to transform fields ranging from artificial intelligence to healthcare. Sheffield will lead one of the centre’s four core technical strands, focusing on computational neuroscience and algorithm development for neuromorphic hardware.
Under the leadership of Professor Eleni Vasilaki from the School of Computer Science, and Head of the Machine Learning Research Group, the Sheffield team is coordinating national efforts to apply principles from biology to reimagine computing systems. Working alongside Professor Tom Hayward, from the School of Chemical, Materials and Biological Engineering/Centre for Machine Intelligence, the team will develop algorithms optimised for neuromorphic hardware — designed to support lifelong learning through speed, efficiency, and resilience.
“We’re not trying to recreate biology,” said Professor Eleni Vasilaki. “We’re identifying which principles make biological systems robust and adaptable — and using those to design computing systems that can learn continuously, adapt in real time, and remain resilient to disruption. It’s about engineering systems that reflect function, not imitation.”
Artificial Intelligence systems often suffer from what’s known as catastrophic forgetting — the tendency to overwrite prior knowledge when learning new tasks. Inspired by how biological brains adapt through selective plasticity and memory replay, Sheffield’s research is set to develop algorithms that enable neuromorphic hardware to learn dynamically over time, without sacrificing speed or efficiency.
These algorithms will be tailored to neuromorphic architectures, capitalising on their unique structure — including massively parallel processing and recurrent connectivity — to offer advantages over traditional systems in both performance and energy use.
By incorporating biological principles of fault tolerance, Sheffield researchers aim to develop algorithms that remain effective even when deployed on imperfect or variable hardware — a crucial requirement for real-world use. They will collaborate with other IKC partners exploring memristive and scalable photonics-based architectures to ensure these algorithms remain compatible, efficient, and robust across diverse platforms.
Professor Tom Hayward said: “Realising new types of computing technologies requires us to bring together scientists and engineers from a wide range of disciplinary backgrounds. It’s incredibly exciting for Sheffield to join a leading national effort to achieve this, and to work towards an age where advanced artificial intelligence is powerful, sustainable, and ubiquitously available.”
Bringing together leading researchers from UCL, Sheffield, King’s College London, Cambridge, Oxford, Manchester, Strathclyde, Imperial College London, and the National Physical Laboratory (NPL), Neuroware will act as a national hub to drive new technologies from lab to market.
Sheffield’s role anchors the centre’s bio-inspired direction, ensuring that biological principles are translated into concrete design decisions — both in how the hardware is built and how it learns.