This proposal seeks funding to create a Centre for Doctoral Training (CDT) in Connected Electronic and Photonic Systems (CEPS). Photonics has moved from a niche industry to being embedded in the majority of deployed systems, ranging from sensing, biophotonics and advanced manufacturing, through communications from the chip-to-chip to transcontinental scale, to display technologies, bringing higher resolution, lower power operation and enabling new ways of human-machine interaction. These advances have set the scene for a major change in commercialisation activity where electronics photonics and wireless converge in a wide range of information, sensing, communications, manufacturing and personal healthcare systems. Currently manufactured systems are realised by combining separately developed photonics, electronic and wireless components. This approach is labour intensive and requires many electrical interconnects as well as optical alignment on the micron scale. Devices are optimised separately and then brought together to meet systems specifications. Such an approach, although it has delivered remarkable results, not least the communications systems upon which the internet depends, limits the benefits that could come from systems-led design and the development of technologies for seamless integration of electronic photonics and wireless systems. To realise such connected systems requires researchers who have not only deep understanding of their specialist area, but also an excellent understanding across the fields of electronic photonics and wireless hardware and software. This proposal seeks to meet this important need, building upon the uniqueness and extent of the UCL and Cambridge research, where research activities are already focussing on higher levels of electronic, photonic and wireless integration; the convergence of wireless and optical communication systems; combined quantum and classical communication systems; the application of THz and optical low-latency connections in data centres; techniques for the low-cost roll-out of optical fibre to replace the copper network; the substitution of many conventional lighting products with photonic light sources and extensive application of photonics in medical diagnostics and personalised medicine. Many of these activities will increasingly rely on more advanced systems integration, and so the proposed CDT includes experts in electronic circuits, wireless systems and software. By drawing these complementary activities together, and building upon initial work towards this goal carried out within our previously funded CDT in Integrated Photonic and Electronic Systems, it is proposed to develop an advanced training programme to equip the next generation of very high calibre doctoral students with the required technical expertise, responsible innovation (RI), commercial and business skills to enable the £90 billion annual turnover UK electronics and photonics industry to create the closely integrated systems of the future. The CEPS CDT will provide a wide range of methods for learning for research students, well beyond that conventionally available, so that they can gain the required skills. In addition to conventional lectures and seminars, for example, there will be bespoke experimental coursework activities, reading clubs, roadmapping activities, responsible innovation (RI) studies, secondments to companies and other research laboratories and business planning courses. Connecting electronic and photonic systems is likely to expand the range of applications into which these technologies are deployed in other key sectors of the economy, such as industrial manufacturing, consumer electronics, data processing, defence, energy, engineering, security and medicine. As a result, a key feature of the CDT will be a developed awareness in its student cohorts of the breadth of opportunity available and the confidence that they can make strong impact thereon.

Soft image sensors are expected to take vital roles in our future daily life. They can monitor the physiological information of our body to provide real-time, noninvasive medical diagnostics, as well as capture and share photos, videos via wireless communications. However, current image sensing electronics cannot be integrated easily into humans, because they are made of rigid semiconductor photodetectors and integrated with optical filters for colour discrimination. In addition, the use of filter creates additional requirements on the optical path difference, which confines the foldability and limits the resolution of the detector array. To overcome these technological limitations, filterless foldable photodetectors which only detect light within a specific wavelength have emerged as critical elements for building soft image sensors. Colloidal quantum dots, metal halide perovskite and organic photodetectors have shown excellent flexibility and detectivity. However, their broad light absorption means filters need to be added to make them specific to a certain colour of light. So far, the most successful filterless model is based on charge collection narrowing (CCN) photodiodes, which are semiconductor devices that convert the specific colour of light into an electrical current. However, since the narrowband response is delivered by controlling photogenerated charge collection efficiency, micrometres thickness junction is often required, which results in an array with a greater likelihood of interpixel cross-talk and frequency bandwidth limitations. It has been demonstrated that the junction thickness can be reduced by using high reflectivity cavities, but a number of challenges still remain. In this research, we aim to tackle these challenges to help find suitable semiconductors that use non-toxic elements and are able to efficiently detect light within a specific wavelength of interest at thicknesses as little as few hundred nanometres. If successful, we would be moving a step closer to an eco-friendly soft image sensor with the potential for many applications. Among all incarnations of solution-processed semiconductors, the recently discovered two-dimensionally (2D) Colloidal Quantum Wells (CQWs) are highly promising for soft image sensor applications, not only do they offer high colour purity with ultranarrow full-width at half-maximum (FWHM) but they also exhibit excellent compatibility with flexible electronics, such as unique stretching enhanced optical polarisation. Unlike colloidal quantum dots, CQW ensembles have no inhomogeneous broadening due to an atomically-precise definition of the short axis and is the reason why CQWs exhibit the narrowest ensemble absorption and emission spectrum of any solution-processed material reported to date. However, looming over much of this success is the fact that all the reported CQWs include toxic heavy metals (e.g., cadmium and lead), and little progress has been made on the fabrication of non-toxic CQWs or CQW narrowband photodetectors. This proposal is therefore designed to substantially address this challenge by using non-toxic mechanically stretchable 2D solution-processed CQWs for the fabrication of soft image sensors. This proposal starts from the growth and surface functionalisation of non-toxic CQWs followed by predictions of the new cavity and charge transport layers for fast CCN. The proposed work will consider the key factors limiting frequency bandwidth, and will demonstrate the inkjet printing of multi-coloured CCN-based photodiodes in a soft image sensor scenario. The high impact objective of this project is the demonstration of a CQWs image sensor which is stretchable and mechanically conformable. This proposal will be underpinned from the established compound semiconductor research expertise at Cardiff University, in close collaboration with Oxford, Cambridge and Bristol University, TCL Corporate Research, Huawei UK, Glaia, 99P Recycling and Hamamatsu UK.

The UKRI Centre for Doctoral Training in Machine Intelligence for Nano-electronic Devices and Systems (MINDS-CDT) will operate as a centre of training excellence in the next generation of systems that employ Artificial Intelligence (AI) algorithms in low-cost/low-power device technologies: hardware-enabled AI. The use of AI in real-world applications through systems of interconnected devices (so-called Internet of Things) is increasingly important across the global economy. Various market surveys estimate the sector to be valued in the hundreds of billions, and project levels of compound annual growth of 25-30%. Applications of these technologies include smart cities, industrial IoT and robotics, connected health and smart homes. It is widely agreed that new advances in artificial intelligence and machine learning are key to unlocking the potential of these systems. Significant challenges remain, however, in the development of robust algorithms and coordinated systems that are efficient, secure, and work in concert with modern devices. Advances in electronics will soon hit atomic scales, requiring new approaches if we are to continue to improve hardware speed and power consumption. Novel nanotechnologies such as memristors have the potential to play a key role in addressing these challenges, but critical to their employment in real-world applications is how algorithms work in the context of device physics. Further, there are significant challenges around how resources available to devices (energy, memory, etc.) can more effectively adapt to the computational tasks at hand, again requiring us to think about how hardware and software work together. The MINDS CDT is unique in its cross-disciplinary research programme crossing emerging AI algorithms and models with advances in device technologies that underpin and enable their potential. To quote from one of our industry partners, "innovation is to come from software and hardware co-development" and that "this joined-up thinking as a potential game changer". The MINDS-CDT will train a substantial number of experts with the knowledge and skills to lead the development of this next generation of intelligent, embedded systems. The training programme will draw from both computer science and electronics expertise at the University of Southampton, and a substantial network of stakeholders from across industry, government and the broader economy. Core to our training ethos is the up-front investigation of the potential impacts of technological innovation on society, security and safety, and in the engagement of interest groups and the public in understanding the benefits as well as the risks of the use of these new developments in AI and technology for our society and economy. The processes we will use here include that all projects and research activities will be informed by in-depth impact assessment, and we will instigate an ambassadors programme for public engagement and, in particular, the engagement of underrepresented groups in AI and engineering.

Over the past years deep learning has brought a revolution in the area of artificial intelligence (AI), producing remarkable results in a variety of application domains including computer vision, natural language processing, speech recgonition, robotics, and clinical decision making. Despite the success of deep learning over a wide spectrum of real-world tasks, there is no doubt that many of the problems that are really at the core of AI are far from being solved. Reasoning, namely taking pieces of information, combining them together, and using these to draw logical conclusions or devise new information, is not a general-purpose capability for modern AI. Imagine an airplane passenger sitting in an exit row, studying the emergency guide, which is often a combination of images and text. Their brain combines visual and textual information in order to infer the intended message -- open the door in the unlikely event of an emergency. A computer system seeing the same document would first employ an image recognition model to scan the image. An Optical Character Recognition (OCR) system would read the text, and a third system would correlate the image and text to understand the complete picture. Although the fundamental principles of analyzing the world around us and the approach a machine takes to process complex information are both based on breaking down the data to its core elements, humans are instinctively better at correlating and integrating information from different modalities, and re-using previously acquired experience and expertise to transfer it to radically different challenges and domains. Today's neural networks fail disastrously when exposed to data outside the distribution they were trained on, overly adhere to superficial and potentially misleading statistical associations instead of learning true causal relations, are unable to reason on an abstract level, which makes it difficult to implement high-level cognitive functions, and are essentially black boxes with with respect to human understanding of their predictions. This fellowship aims to alleviate these deficiencies by developing a new class of neural network models which will demonstrate reasoning capabilities, a skill required to enhance many AI applications. Rather than relying on a monolithic network structure, we propose to assemble a network from a collection of more specialized modules, making use of an explicit, modular reasoning process, which allows for differentiable training (with backpropagation) but without expert supervision of reasoning steps. We will develop a theoretical framework which characterizes what it means for neural network models to reason, design various reasoning modules, and showcase their practical importance in applications which understand requests and act on them, process and aggregate large amounts of data (e.g., from multiple modalities), make generalizations (e.g., robots cannot be pretrained on all possible scenarios they might encounter), deal with changing situations and causality, manifest creativity (e.g., in writing a story or a poem), co-ordinate various agents (e.g., in game playing), and are able explain their predictions and decisions. The proposed Fellowship will have a transformative effect on AI theory and practice. It sets an ambitious agenda which unifies multiple strands of AI research, bridging the gap between the neural and symbolic views of AI and integrating their complementary strengths. It will provide the means for developing a UK skill base in AI, and wil have wide ranging impact in academia, industry, the UK economy, and society e.g., , by embedding AI in many domains of daily life and rendering tools such as neural networks more explainable.

We propose the development of a new technology for Non-Invasive Single Neuron Electrical Monitoring (NISNEM). Current non-invasive neuroimaging techniques including electroencephalography (EEG), magnetoencephalography (MEG) or functional magnetic resonance imaging (fMRI) provide indirect measures of the activity of large populations of neurons in the brain. However, it is becoming apparent that information at the single neuron level may be critical for understanding, diagnosing, and treating increasingly prevalent neurological conditions, such as stroke and dementia. Current methods to record single neuron activity are invasive - they require surgical implants. Implanted electrodes risk damage to the neural tissue and/or foreign body reaction that limit long-term stability. Understandably, this approach is not chosen by many patients; in fact, implanted electrode technologies are limited to animal preparations or tests on a handful of patients worldwide. Measuring single neuron activity non-invasively will transform how neurological conditions are diagnosed, monitored, and treated as well as pave the way for the broad adoption of neurotechnologies in healthcare. We propose the development of NISNEM by pushing frontier engineering research in electrode technology, ultra-low-noise electronics, and advanced signal processing, iteratively validated during extensive tests in pre-clinical trials. We will design and manufacture arrays of dry electrodes to be mounted on the skin with an ultra-high density of recording points. By aggressive miniaturization, we will develop microelectronics chips to record from thousands of channels with beyond state-of-art noise performance. We will devise breakthrough developments in unsupervised blind source identification of the activity of tens to hundreds of neurons from tens of thousands of recordings. This research will be supported by iterative pre-clinical studies in humans and animals, which will be essential for defining requirements and refining designs. We intend to demonstrate the feasibility of the NISNEM technology and its potential to become a routine clinical tool that transforms all aspects of healthcare. In particular, we expect it to drastically improve how neurological diseases are managed. Given that they are a massive burden and limit the quality of life of millions of patients and their families, the impact of NISNEM could be almost unprecedented. We envision the NISNEM technology to be adopted on a routine clinical basis for: 1) diagnostics (epilepsy, tremor, dementia); 2) monitoring (stroke, spinal cord injury, ageing); 3) intervention (closed-loop modulation of brain activity); 4) advancing our understanding of the nervous system (identifying pathological changes); and 5) development of neural interfaces for communication (Brain-Computer Interfaces for locked-in patients), control of (neuro)prosthetics, or replacement of a "missing sense" (e.g., auditory prosthetics). Moreover, by accurately detecting the patient's intent, this technology could be used to drive neural plasticity -the brain's ability to reorganize itself-, potentially enabling cures for currently incurable disorders such as stroke, spinal cord injury, or Parkinson's disease. NISNEM also provides the opportunity to extend treatment from the hospital to the home. For example, rehabilitation after a stroke occurs mainly in hospitals and for a limited period of time; home rehabilitation is absent. NISNEM could provide continuous rehabilitation at home through the use of therapeutic technologies. The neural engineering, neuroscience and clinical neurology communities will all greatly benefit from this radically new perspective and complementary knowledge base. NISNEM will foster a revolution in neurosciences and neurotechnology, strongly impacting these large academic communities and the clinical sector. Even more importantly, if successful, it will improve the life of millions of patients and their relatives