A 25-$μ$s/inf Event-driven Graph Neural Network Processor with Spatiotemporal Caching and Spline Convolution for Ultra-low-latency AI at the Edge

2026-09-14

AI Summary

本文提出了一种基于时空缓存和样条卷积的事件驱动图神经网络处理器ETHEREAL，解决了边缘计算中超低延迟AI处理的问题。

Abstract

Dynamic-vision-sensor (DVS) cameras generate events on a per-pixel basis with a $μ$s-level temporal resolution, calling for new algorithm-hardware co-design approaches compared to standard frame-based vision. While event-driven graph neural networks (EV-GNNs) emerge as a promising algorithmic solution, they raise new HW challenges by mixing dense-regular compute operations and sparse-irregular memory accesses. We present ETHEREAL, the first EV-GNN accelerator that scales to 640$\times$480 resolutions, thanks to a neighbor-parallel spline convolution engine and a 2D/3D-split memory hierarchy with a novel region-of-interest spatiotemporal caching mechanism. Measurement results demonstrate end-to-end inference with 25.6$μ$s latency and 1.7$μ$J energy per event on state-of-the-art workloads

Problem

• event-driven
• graph neural network
• low-latency AI
• hardware challenge
• dynamic vision sensor

Innovation

• event-driven graph neural networks
• spline convolution
• spatiotemporal caching