一句话：提出 Hilbert 曲线初始放置与力导向优化的 SNN 映射方法，支持 40 亿神经元并降低能耗、延迟与网络拥塞。
神经形态硬件是面向脉冲神经网络（SNN）的多核系统。随着硬件规模增长，将大规模 SNN 高效映射到硬件十分困难。本文提出一种超大规模 SNN 映射方法，旨在降低能耗、脉冲延迟和片上网络通信拥塞。方法分两步：先用具有空间填充特性的 Hilbert 曲线生成初始放置；再用力导向（FD）算法优化，将簇间连接建模为张力，把局部放置优化转化为受力分析问题。评估规模达 40 亿神经元，比此前研究大 200 倍以上。结果表明该方法达到当前最优性能，显著优于已有方法。
Neuromorphic hardware is a multi-core computer system specifically designed to run Spiking Neuron Network (SNN) applications. As the scale of neuromorphic hardware increases, it becomes very challenging to efficiently map a large SNN to hardware. In this paper, we proposed an efficient approach to map very large scale SNN applications to neuromorphic hardware, aiming to reduce energy consumption, spike latency, and on-chip network communication congestion. The approach consists of two steps. Firstly, it solves the initial placement using the Hilbert curve, a space-filling curve with unique properties that are particularly suitable for mapping SNNs. Secondly, the Force Directed (FD) algorithm is developed to optimize the initial placement. The FD algorithm formulates the connections of clusters as tension forces, thus converts the local optimization of placement as a force analysis problem. The proposed approach is evaluated with the scale of 4 billion neurons, which is more than 200 times larger than previous research. The results show that our approach achieves state-of-the-art performance, significantly exceeding existing approaches.
@inproceedings{paperbot10079,
title = {Mapping Very Large Scale Spiking Neuron Network to Neuromorphic Hardware},
author = {Ouwen Jin and Qinghui Xing and Ying Li and Shuiguang Deng and Shuibing He and Gang Pan},
booktitle = {ASPLOS},
year = {2023},
doi = {10.1145/3582016.3582038}
}