EPSRC logo

Details of Grant 

EPSRC Reference: EP/Z534407/1
Title: Neumat Network: Neuromorphic Materials and Devices for Future AI Hardware
Principal Investigator: Driscoll, Professor JL
Other Investigators:
Mehonic, Dr A Serb, Dr A
Researcher Co-Investigators:
Project Partners:
Intel Corporation Ltd Royal Academy of Engineering sureCore Ltd
University of Leeds University of Modena and Reggio Emilia University of Oxford
University of Sheffield
Department: Materials Science & Metallurgy
Organisation: University of Cambridge
Scheme: Network TFS
Starts: 27 December 2024 Ends: 26 December 2029 Value (£): 1,188,914
EPSRC Research Topic Classifications:
Artificial Intelligence Microsystems
EPSRC Industrial Sector Classifications:
Related Grants:
Panel History:  
Summary on Grant Application Form
The influx of both structured and unstructured data has been meteoric, driven by industries like autonomous vehicles, robotics, IoT, medical technology, security, and entertainment. This wave results in an ever-growing demand for computing power, which doubles nearly every two to three months. Artificial Intelligence (AI) stands as the main catalyst for this surge, as its operations consume tremendous energy. Thus, the expansion of AI is producing impending hurdles: the current rate of scaling is unsustainable. As AI demands more computing power, energy consumption surges, especially in data centres. The future of AI hinges on developing technologies that are simultaneously more energy-efficient and more powerful computationally. As a result, data centres now account for an estimated 3% of the world's electricity use, leaving a sizeable carbon footprint.

AI algorithms, such as those used in self-driving vehicles, process vast volumes of data, particularly during training phases. The path forward demands fundamental innovations, starting from the basic hardware and then moving up to the computing which controls the hardware. Beyond vastly improving on today's CMOS technology, there's potential in delving into analogue electronics or adopting neuromorphic (akin to the human brain) approaches.



The UK boasts robust expertise across various facets of this overarching arena, encompassing novel materials, device development, circuitry design, and pioneering AI algorithms. Adopting non-von Neumann computer architectures, namely more brain-like machines, is a very promising way forward for more efficient computing (the brain is a million times more efficient than current computing). The shift to this new hardware mandates a collective effort spanning multiple scientific and engineering fields. To this end, a more cohesive alignment of different sectors is imperative to pioneer breakthroughs that can rival and eventually replace today's prevalent CMOS and transistor-based digital systems. This underscores the need for a concentrated UK initiative, harmonizing expertise from materials science, applied physics, device engineering, circuit design and algorithmic development.



Enter NeuMat: our network designed to be a platform for experts from these various fields, with a main emphasis being on the starting point for revolutionary success: revolutionary hardware. Thus NeuMat aims to catalyse groundbreaking work in the UK on innovative neuromorphic AI hardware technologies. It will foster the exchange of ideas, offer training, facilitate researcher exchanges, share methodologies, build industrial partnerships, map out future directions, and most importantly build a strong cohort of early career researchers to carry this field forward in the UK in the future.

NeuMat will cultivate a cohesive network comprising academics, industry specialists, and importantly also PhD students and postdocs. By doing so, we aim to bolster academic-industrial partnerships and catalyse the development of innovative industry products. Our early-career project leaders are poised to initiate a subsequent network project, ensuring sustained momentum. Upon the project's conclusion, we'll craft a detailed roadmap to serve as a directive for UK policymakers navigating this rapidly evolving and crucial research sector. We will also be self-sustaining and put firm plans in place for a follow-on-network on a related area which emerges as the most timely topic at the end of this network.
Key Findings
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Potential use in non-academic contexts
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Impacts
Description This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Summary
Date Materialised
Sectors submitted by the Researcher
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Project URL:  
Further Information:  
Organisation Website: http://www.cam.ac.uk