Neuromorphic Computing and AI for Energy-Efficient and Adaptive Edge Intelligence

Artificial intelligence (AI) has become a pivotal field in scientific research, yet its progress has been limited by reliance on massive datasets, high computational demands, and energy inefficiency. The growing need for real-time AI processing on edge devices—from autonomous vehicles to wearables—further exposes these limitations, particularly in terms of latency, as traditional AI systems struggle with power, memory, and responsiveness under constrained environments. Neuromorphic computing, inspired by the brain’s efficiency, offers a transformative solution by enabling parallel, adaptive learning with ultra-low power consumption, making it ideal for edge computing and real-time applications.

This Research Topic focuses on addressing the challenges of operational efficiency and resource constraints in edge environments through cutting-edge methods that integrate neuromorphic computing and artificial intelligence. The core objective is to draw inspiration from the structure and dynamics of biological neural systems to develop corresponding computational models, next-generation neuromorphic chip architectures, and application-driven algorithms. This Research Topic aims to promote a unified perspective on how neuromorphic computing can enable the next wave of intelligent edge systems.

We emphasize a holistic approach that encompasses hardware design, spike-based computation models, adaptive learning algorithms, and supporting software frameworks. Particular focus will be placed on energy-efficient, event-driven systems with broad applicability across real-world scenarios. Contributions demonstrating practical implementations in edge intelligence, distributed sensor networks, and real-time control are especially encouraged. By bridging neurobiological principles with engineering innovation, this initiative aims to advance biologically plausible, sustainable AI systems that transcend the limitations of traditional von Neumann architectures.

Topics of interest include, but are not limited to:

1. Novel Brain-inspired computational frameworks for edge systems (e.g. modeling neural dynamics, synaptic plasticity, and event-driven processing mechanisms)

2. Neuromorphic chip architectures for edge systems (e.g. in-memory computing/spiking neuron implementations)

3. Neuromorphic hardware design and implementation for edge systems.

4. Spiking neural network optimization and deployment on edge platforms.

5. Neuromorphic computing algorithms or event-driven learning rules for edge devices.