Overview
INCC 2027 invites original abstract submissions across the full spectrum of neuromorphic and unconventional computing. Organised by NeuroSYNC — the UK Multidisciplinary Centre for Neuromorphic Systems and Computing, the conference brings together researchers, industry leaders, innovators, and policymakers to explore and shape the future of brain-inspired and emerging physics-based computing technologies.
Spanning neuroscience, materials, devices, hardware, concepts and algorithms, and real-world applications, INCC 2027 recognises that major advances often emerge at the interfaces between disciplines. Rather than being organised as separate tracks, the conference is built around five interconnected and flexible streams that explore common scientific challenges from different perspectives — spanning physics, materials, new concepts and algorithms, hardware, systems, and applications. The themes below are intended to guide the scope of the conference rather than define its limits, and contributions on related and emerging topics are encouraged.
Neuromorphic and unconventional computing bring together ideas from many disciplines. Reflecting this, INCC 2027 is organised into five interconnected and flexible streams that explore shared scientific challenges from different perspectives — spanning neuroscience, physics, materials, algorithms, hardware, and real-world applications. The listed themes are intended to guide the scope of the conference rather than define its limits, and contributions on related topics are welcome.
Topic of Interests
Stream 1: Neuroscience-Inspired Unconventional Computing & Bio-Electronics Interfaces
- Hybrid bio–electronic neural systems — bio-silicon integration, closed-loop interfaces, living neural systems
- Neuromorphic sensing and event-based perception — event-based encoding and sensor fusion, vision, multimodal sensing
- Dendritic and synaptic computation — multi-compartment neural models, nonlinear processing, dendrocentric learning
Plasticity and adaptive learning in hardware — spike-timing-dependent plasticity, multi-timescale learning, lifelong learning, attractors and memory formation - Organoid and biological computing platforms — organoid intelligence, functional network formation, plasticity in organoids                                                                                                                                         Â
Stream 2: Physical Computing: Concepts & Algorithms
- Reservoir and dynamical computing — reservoir computing, extreme learning machines, dynamical regimes, nonlinearities, computing capacity, multi-scale topologies, and deep reservoir constraints.
- Unconventional computing — probabilistic, stochastic, energy-based, topological, quantum, over-the-air, hyperdimensional and symbolic computing approaches.
- Learning in physical systems — physical annealing, gradient propagation, digital twins, physics-aware training, and model-/gradient-free learning.
Performance theory — predictors of computing performance, dynamical regimes, topologies and scaling laws. - Physics-informed and hybrid AI — physics-informed computing, symbolic reasoning, neuromorphic systems, vector symbolic architectures, and parameter scaling.
- Algorithm–hardware co-design — co-optimisation, readout mechanisms and hardware-efficient learning methods.        Â
Stream 3: Materials for Computing
- Synapses and in-materio learning — bio-plausible plasticity, multi-timescale plasticity
- Nonlinear and functional materials for computation — low-energy nonlinearity, activation-function-like nonlinearities, ultra-fast nonlinear responses
- Memristive and resistive switching materials — phase-change materials, ferroelectric, semiconducting and multiferroic materials
- 2D materials and van der Waals heterostructures
- Spintronic materials and magnetic devices
- Optoelectronic and photonic materials — oxides, dielectrics, semiconductors, organic bio-compatible and soft materials
- Stochastic and disorder-based materials for computing                                                                                                                                                                                                            Â
Stream 4: Neuromorphic Computing Hardware & Full System Implementation
- Reservoir and dynamical computing — reservoir computing, extreme learning machines, dynamical regimes, nonlinearities, computing capacity, multi-scale topologies, and deep reservoir constraints.
- Unconventional computing — probabilistic, stochastic, energy-based, topological, quantum, over-the-air, hyperdimensional and symbolic computing approaches.
- Learning in physical systems — physical annealing, gradient propagation, digital twins, physics-aware training, and model-/gradient-free learning.
Performance theory — predictors of computing performance, dynamical regimes, topologies and scaling laws. - Physics-informed and hybrid AI — physics-informed computing, symbolic reasoning, neuromorphic systems, vector symbolic architectures, and parameter scaling.
- Algorithm–hardware co-design — co-optimisation, readout mechanisms and hardware-efficient learning methods.
Stream 5: Applications, Industry, Policies & Technology Transfer
- Commercial neuromorphic processors and deployment
- Edge neuromorphic hardware for robotics and autonomous systems
- Sensing, IoT and cyber-physical systems
- Neuromorphic technologies for critical infrastructure and supply chain resilience
- Large-scale computing, sustainability and energy efficiency
- Standardisation, industrial benchmarking and interoperability of neuromorphic hardware
- Ethical, regulatory and societal aspects of neuromorphic AI
- Policies, industrial roadmaps and global international research initiatives
Abstract Submission Guidelines
Format
- Maximum 2 pages A4, including figures and references
- Submit as a Word document (.docx) — not PDF
- Do not deviate from the preset formatting in the official template
- First line of each subsequent paragraph within sections indented by 0.5 cm
- Sections separated by a single blank line
Required Structure
- Title
- Authors & Affiliations — name, department, institution, city, country; corresponding author email marked with *
- 50-word Summary — main motivation and key results of the presentation
- I. Introduction — motivation, state-of-the-art review, brief description of the work (3–5 references recommended)
- II. Methods & Results — experimental setup, device/simulation details, main results, conclusions, and wider significance; this section should constitute the bulk of the abstract
- III. References — standard journal citation format, limited to 10 references
- Figure — maximum 1 figure; single row, up to 2 panels; remove table borders before submission; no more than 3 lines of caption
Awards
Awarded to the best paper presented at INCC 2027.
After Acceptance
An abstract ID will be assigned for identification and inclusion in the conference abstract booklet.
Review Process
All submitted abstracts will be reviewed by the relevant Programme Subcommittee. Abstracts will be assessed on scientific novelty, clarity of presentation, and relevance to the interdisciplinary scope of INCC 2027.
Best Student Paper Awards
Awarded to the best papers with a student as first author.
Programme Chairs
Each of the five streams is overseen by a dedicated subcommittee. For the full list of subcommittee members, please visit the Programme page on the conference website.
•   Stream 1: Paul Roach (Loughborough University)
•   Stream 2: Natalia Berloff (University of Cambridge)
•   Stream 3: Judith Driscoll (University of Cambridge)
•   Stream 4: Tony Kenyon (University College London)
•   Stream 5: Bert Offrein (IBM)