Prosperity：通过乘积稀疏性加速脉冲神经网络
Prosperity: Accelerating Spiking Neural Networks via Product Sparsity
AI总结：
本文提出名为乘积稀疏性的新型稀疏范式，利用矩阵乘法中的组合相似性复用内积结果以减少冗余计算，并设计了Prosperity架构实时消除冗余，在SpikeBERT模型上实现11倍计算减少，相比PTB和A100 GPU分别获得7.4倍和1.8倍加速。
摘要：
脉冲神经网络（SNNs）因其基于脉冲的激活方式而具有极高的效率，这种方式本质上会产生比特稀疏的计算模式。现有的SNN硬件实现利用这种稀疏模式来避免浪费的零值计算，然而这种方法未能充分利用SNN的潜在效率。本研究引入了一种名为乘积稀疏性（Product Sparsity）的新型稀疏范式，它利用矩阵乘法运算内部的组合相似性来复用内积结果并减少冗余计算。与传统的比特稀疏方法相比，乘积稀疏性在不影响原始计算结果的情况下显著提升了SNN的稀疏度。例如，在SpikeBERT SNN模型中，乘积稀疏性实现了仅1.23%的密度并减少了11倍的计算量，而比特稀疏性的密度为13.19%。为了高效实现乘积稀疏性，我们提出了Prosperity架构，该架构解决了实时识别并消除冗余计算的挑战。与先前的SNN加速器PTB和A100 GPU相比，Prosperity分别实现了平均7.4倍和1.8倍的加速，以及8.0倍和193倍的能效提升。Prosperity的代码可在https://github.com/dubcyfor3/Prosperity获取。
英文摘要：
Spiking Neural Networks (SNNs) are highly efficient due to their spike-based activation, which inherently produces bit-sparse computation patterns. Existing hardware implementations of SNNs leverage this sparsity pattern to avoid wasteful zero-value computations, yet this approach fails to fully capitalize on the potential efficiency of SNNs. This study introduces a novel sparsity paradigm called Product Sparsity, which leverages combinatorial similarities within matrix multiplication operations to reuse the inner product result and reduce redundant computations. Product Sparsity significantly enhances sparsity in SNNs without compromising the original computation results compared to traditional bit sparsity methods. For instance, in the SpikeBERT SNN model, Product Sparsity achieves a density of only $1.23\%$ and reduces computation by $11\times$, compared to bit sparsity, which has a density of $13.19\%$. To efficiently implement Product Sparsity, we propose Prosperity, an architecture that addresses the challenges of identifying and eliminating redundant computations in real-time. Compared to prior SNN accelerator PTB and the A100 GPU, Prosperity achieves an average speedup of $7.4\times$ and $1.8\times$, respectively, along with energy efficiency improvements of $8.0\times$ and $193\times$, respectively. The code for Prosperity is available at https://github.com/dubcyfor3/Prosperity.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
