A Reconfigurable Neuromorphic Compute System for Brain-Scale Simulations

该项目旨在构建世界上第一个可重构的神经形态计算系统。该硬件旨在高效运行大脑规模的模拟，为我们提供一个平台来发展我们对大脑的理解，并开发大脑规模的计算应用程序。预期的结果是能够有效地模拟生物大脑，用于计算神经科学研究和为实际应用调查新的机器学习方法。澳大利亚和全球在神经科学和机器学习方面的研究社区将受益于该基础设施，因为它可以通过互联网远程访问，将全球范围的协作研究释放到脑级计算中。

The project aims to construct a world-first reconfigurable neuromorphic compute system. The hardware is designed to run brain-scale simulations efficiently, providing a platform to develop our understanding of the brain and develop brain-scale computing applications. Expected outcomes are to enable the efficient simulation of biological brains for computational neuroscience research and investigation of novel machine learning approaches for practical applications. The Australian and global research community in neuroscience and machine learning would benefit from the infrastructure, as it can be accessed remotely via the internet, unlocking world-wide collaborative research into brain-scale computing.