1. Summary of the context and overall objectives of POST-DIGITAL
Having closed in 2024, POST-DIGITAL was a 4-year European Training Network (ETN) with 15 ESRs, focusing on state-of-the-art non-digital computing in unconventional substrates, in a radical departure from classical digital computing, training its PhD students in the inter-disciplinary fields of emerging disruptive neuromorphic computational technologies and their applications.
Coordinated by Aston University (UK), the consortium united world-leading academic and industrial groups in disciplines such as machine learning, computer science, physical neural networks, reservoir computing, signal processing, optical communications, photonic implementation of computing systems, unconventional neuromorphic circuit and chip design, and fiber-optic technology.
The strong industrial presence in the network provided ESRs with experience of practical applications and solutions beyond traditional digital methods, allowing them to develop into a new generation of scientific and industrial leaders, strengthening Europe’s human resources and industry competitiveness in the future post digital economy and technology.
Our vision is that neuromorphic, brain- and nature-inspired technologies offer substantial advantages in terms of processing capabilities and power efficiency. We are confident that the project’s outcomes, including delivering 15 highly trained researchers with strong potential to become the next generation of academic/industrial leaders, will bring about benefits for the development of faster processing, significantly higher bandwidth efficiency and adaptability through integration of self-learning systems.
POST-DIGITAL and its ESRs have already made significant impact and are expected to continue to do so after POST-DIGITAL has ended:
-
published 90+ dissemination activities (peer reviewed journal and conference papers; talks and posters in workshop and other meetings)
-
gave 40+ presentations at prestigious international conferences
-
organised 27 scientific training events attended by 1150 external researchers
-
delivered 27 scientific outreach activities
-
completed 70+ months of inter-sectorial secondment
-
had a research proposal as Co- PI accepted
-
co-invented a US Patent US11574178B2: ‘Method and system for machine learning using optical data’
-
co-founded the AI Development Platform Adaptive ML, and raised $20M seed investment
-
secured influential research internship at Google
-
received promotions in industry
2. Work performed and main results achieved
The core of the scientific work was performed in WP1-WP4 which is summarised below, while WP5-WP8 covered Recruitment/Management/Governance; Training; Dissemination/Exploitation and Ethics.
WP1: NEW CONCEPTS AND THEORY
There is an urgent need for developing a unified theory of non-digital computing, to integrate the multiplicity of the as yet largely disconnected research in different traditional disciplines. In WP1, a range of architectures, algorithms and analyses were realized by ESRs. These gave new insights into our theoretical understanding of information processing in non-digital physical dynamical systems. The contributions include analyses of information encodings in recurrent neural networks with methods from dynamical systems theory. Deliverables D1.2 and D4.2 give an overview coverage of these lines of work, which was realized in the (often collaborative) PhD projects of the ESRs.
WP2: IMPLEMENTATION AND CHARACTERIZATION
Large-scale photonic neural networks have been designed and tested by several ESRs. This has been possible thanks to the efforts regarding multiplexing strategies and scalability. Photonic neural networks based on optoelectronic systems with delay (time multiplexing), frequency combs (frequency multiplexing), and large-area lasers (space multiplexing) have been demonstrated. Key achievements of WP2 include parallelizing and cascading information processing, and implementing high bandwidth (GHz) systems. Interaction with other WPs has been fruitful. WP2 has successfully implemented concepts designed with WP1 and has significantly improved performance in classification, prediction, and inference tasks. In addition, WP2 has provided design guidelines for photonic integrated circuits demonstrated in WP3. Finally, the promising applications proposed in WP4 have been enabled by the systems developed in WP2.
WP3: INTEGRATED SYSTEMS
In WP3, ESRs have investigated different reservoir architectures (spatial multiplexing, crossbar arrays, frequency multiplexing, etc), both theoretically and experimentally. These have been applied to a number of telecom tasks, mostly dispersion and nonlinearity compensation for different modulation formats (IM, self-coherent, Kramers-Kronig…). Different prototypes have been fabricated and characterized, showing a.o. online learning.
WP4: BENCHMARKING AND APPLICATIONS
WP4 was focused on benchmarking neuromorphic hardware, developing novel applications, and improving hardware performance. Key research objectives included analysing the performance in practical environments, identifying industrial applications, and fostering industry collaborations. The benchmarking aspect of WP4 involved comparing novel neuromorphic solutions with conventional digital solutions. The benchmarking evaluated different computing paradigms, node counts, processing speeds, energy consumption, and footprints. 
The focus was on analog physical computing and photonic substrates, comparing implementations of reservoir computing, extreme learning machines, and feedforward neural networks.
Detailed results from WP1-4 have been published Open Access (accessed via https://cordis.europa.eu/project/id/860360/results). A final collaborative review paper on ‘A Perspective on Computing with Physical Substrates’ written by the entire POST-DIGITAL consortium has been accepted for publication in the high-impact journal ‘Reviews in Physics’.
3. Summary of individual ESRs’ work
4. Progress beyond the state of the art and potential impacts
POST-DIGITAL made multiple contributions to the field beyond the state of the art. This includes gaining deeper insights into the fundamental understanding of information processing in non-digital real-world dynamical systems. POSTDIGITAL explored a range of architectures, algorithms and concepts contributing to the mathematical analyses of information encoding in recurrent neural networks with methods from dynamical systems theory. New types of large-scale photonic neural networks have been designed and tested, offering parallelization of information processing, cascadability and high speed (GHz bandwidth) implementations. Promising strategies for scalability have been developed and demonstrated in photonic neural networks based on systems with delay (time multiplexing) , using frequency combs (frequency multiplexing), and large-area lasers (space multiplexing). POSTDIGITAL contributed to the progress in design and fabrication of integrated versions of neuromorphic systems and different reservoir computing architectures, implementing spatial multiplexing, crossbar arrays, frequency multiplexing and more. The realized systems have been tested in real-world applications such as dispersion and nonlinearity compensation in optical communications for different modulation formats. POSTDIGITAL industrial partners ensured benchmarking of the developed neuromorphic hardware by comparing novel photonic implementations with conventional solutions. Overall, POSTDIGITAL created an important platform for non-digital technologies of the future and trained 15 ESRs in this emerging strategic field.