Memristive Materials for Spike-based Computing

In the era of big data and the Internet-of-Things, a vast amount of sensing data, such as pictures, speeches, and videos, need to be processed in real-time with high energy efficiency. This poses a significant challenge to traditional computing architecture due to the von-Neumann bottleneck. Emulating the brain's architecture and working mechanisms to build a high-efficient intelligent machine has attracted considerable attention and becomes a research hotspot. Emerging devices have shown great potential for emulating spiking neurons and plastic synapses, due to their simple structures, high integration capacity, low power, and rich dynamics etc. For the past few years, the synapse array based on these devices have been widely studied for building artificial neural networks for accelerating the computing of vector-matrix multiplication.

With the unprecedented development of technology, computing needs are pervasive and entrenched in our daily life. In the IoT age, we are witnessing a strong urge to design ultra-low-power intelligent sensory systems that can process the information they acquire. Animal brains are an existence proof of such kind of sophisticated architecture. Due to the von-Neumann bottleneck, the existing computing architectures are incapable to match such performance. Hence emerging materials and technologies are at the centre of attraction.

Over the decade, several emerging devices are showing promise in emerging computing applications. In general, the information is processed with the analog type, there is still a big challenge to pursue the spike-based computing mode that in biological systems. This topic will explore the emerging devices for building spike-based neuromorphic machines, including but not limited to the following researches: the optimization and integration of emerging devices for synapses, the neural circuits design using emerging devices, brain-inspired algorithms implementation based on emerging devices, spiking neural networks implementation, brain-inspired architecture design, the co-optimization of devices, circuits, and algorithms, etc.

· Optimization and integration of emerging devices for spiking synapses

· Neural spiking circuits design using emerging devices

· Brain-inspired spiking algorithms implementation based on emerging devices

· Quantum materials for neuromorphic spiking computing

· High-performance computing for neuroscience simulations

· Signal processing in bioelectronics

· Robust spiking pattern recognition

· Neuro-robotics spiking systems