With both knowing more and more details about how neurons and complex neural networks work and having serious demand for making performable huge artificial networks, more and more efforts are devoted to build both hardware and/or software simulators and supercomputers targeting artificial intelligence applications, demanding an exponentially increasing amount of computing capacity. However, the inherently parallel operation of the neural networks is mostly simulated deploying inherently sequential (or in the best case: sequential–parallel) computing elements. The paper shows that neural network simulators, (both software and hardware ones), akin to all other sequential–parallel computing systems, have computing performance limitation due to deploying clock-driven electronic circuits, the 70-year old computing paradigm and Amdahl’s Law about parallelized computing systems. The findings explain the limitations/saturation experienced in former studies.

中文翻译：

---

阿姆达尔定律如何限制大型人工神经网络的性能：为什么在基于处理器的模拟器上进行全尺寸脑部模拟的功能受到限制。

越来越多的人了解神经元和复杂的神经网络的工作原理，并且迫切需要建立可运行的巨大人工网络，因此，越来越多的工作致力于构建针对人工智能应用的硬件和/或软件模拟器以及超级计算机，计算能力成倍增加。但是，神经网络的固有并行操作主要是模拟部署固有的顺序（或最佳情况：顺序-并行）计算元素。该论文表明，与所有其他顺序并行计算系统一样，神经网络模拟器（包括软件和硬件）也由于部署了时钟驱动的电子电路而受到了计算性能的限制，已有70年历史的计算范例以及关于并行化计算系统的阿姆达尔定律。这些发现解释了先前研究中的局限性/饱和性。