Non-uniform Memory Partitioning For Low-Power Spiking Neural Networks

Abstract

Spiking Neural Networks (SNNs) naturally excel in processing temporally rich and sparse data. However, because of their time-stepped processing, memory access, specifically to synaptic weights stored in SRAM (static random-access memory), tends to dominate total power consumption. To address this issue, without incurring a large area overhead, we propose to leverage the greatly varying average firing rate of neurons in the network to efficiently allocate synaptic weights to an on-chip memory consisting of multiple non-uniformly sized memory banks. By assigning weights of frequently firing neurons to shallow, low-access cost memory and less actively accessed weights to deeper, high-density memories, the average power consumption of the synaptic weight memory is decreased without incurring a large area overhead. To benchmark our proposed architecture and find optimal configurations of memory arrangements, we perform an automatic exploration based on application requirements and hardware constraints. For memory designs synthesized in 28-nm CMOS technology, we show that our architecture can achieve a synaptic weight memory access power reduction of up to 61\% compared to a conventional design, with a 2.1$\times$ lower area overhead, as compared to a traditional uniformly partitioned memory bank that achieves a comparable reduction.

arXiv comment5 pages, 3 figures, Publishes at IEEE AICAS 2025

References in corpus (4)

• A scalable multi-core architecture with heterogeneous memory structures for Dynamic Neuromorphic Asynchronous Processors (DYNAPs)
• A 0.086-mm${}^2$ 12.7-pJ/SOP 64k-Synapse 256-Neuron Online-Learning Digital Spiking Neuromorphic Processor in 28nm CMOS
• The Heidelberg spiking datasets for the systematic evaluation of spiking neural networks
• Bottom-up and top-down approaches for the design of neuromorphic processing systems: Tradeoffs and synergies between natural and artificial intelligence