LowPASS: A Low power PIM-based accelerator with Speculative Scheme for SNNs
Fangxin Liu, Shiyuan Huang, Longyu Zhao, Li Jiang, Zongwu Wang · 2024
Spiking neural networks (SNNs) are considered as energy-efficient alternatives to deep neural networks (DNNs). By adopting event-driven information processing, SNNs can significantly reduce the computational demands associated with DNNs, while still achieving comparable performance. However, current SNNs primarily prioritize high accuracy and large sparsity by constructing complex neuron models that generate sparse spikes. Unfortunately, this approach results in low energy efficiency and high latency, posing a significant challenge for deploying SNNs at the edge. Furthermore, the dominant computation in SNNs, which involves spike-wise Add-Accumulate operations, is well-suited for process-in-memory (PIM) architectures. However, exploiting high parallel processing and spike sparsity in PIM-based SNN accelerators is challenging due to the irregularity and time dependency of spikes.