Neuromorphic Sensing and Computing Architecture for Next-Gen AI Hardware

– 19–20 January 2024

The workshop was a hub for exploring neuromorphic engineering and brain-inspired systems shaping artificial intelligence hardware. Scholars, researchers, and industry leaders delved into discussions on large-scale neuromorphic computing through talks and tutorial sessions. Both theory and practical applications reshaping artificial intelligence hardware were covered from algorithm to hardware design with practical demos.

The different talks delivered were on (i) Neuromorphic Sensing and Computing at the International Centre for Neuromorphic Systems, (ii) Continual Learning Systems with Neural Plasticity, (iii) A Sampling Theoretic Framework for Neuromorphic Sensing, (iv) Quantum Tunneling, Synaptic Intelligence and Learning-in-Memory, (v) On-Sensor Computer Vision with Pixel-Parallel Processor Arrays, (vi) A distributional approach to Datafree pruning, and (vii) Processing in interconnect inspired by dendritic computation. The tutorial on the second day delved into an in-depth look at the SCAMP chip architecture, instruction set, microarchitecture and circuit implementation, following by its programming and familiarisation on the simulator and GUI.

The workshop on neuromorphic engineering and brain-inspired systems brought together scholars, researchers, and industry leaders for a fruitful exploration of the future of AI hardware. Through a series of talks, tutorials, and practical demonstrations, participants delved into the theoretical foundations and real-world applications of large-scale neuromorphic computing. The second day offered a practical component through a tutorial on the SCAMP chip architecture. Overall, the workshop fostered a collaborative environment for exploring the exciting possibilities of neuromorphic engineering. By bridging theory and practice, the event equipped participants with the knowledge and tools to shape the future of AI hardware.