A scalable time-based integrate-and-fire neuromorphic core with brain-inspired leak and local lateral inhibition capabilities
A scalable time-based integrate-and-fire neuromorphic core with brain-inspired leak and local lateral inhibition capabilities
复制标题
具有受大脑启发的泄漏和局部横向抑制功能的可扩展的基于时间的集成和激发神经形态核心
DOI：
10.1109/cicc.2017.7993627
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
C. Kim
中科院分区：
文献类型：
--
作者：
Muqing Liu;L. Everson;C. Kim
完全可扩展的轻量级集成和发射神经形态核心具有受大脑启发的泄漏和局部侧向抑制功能，在 65nm 中实现。核心使用标准数字电路完全在时域中计算神经网络算法。使用所提出的核心实现的并行两层架构实现了 91% 的手写数字识别准确率。 0.24mm2 神经形态核心包括 64 个数控振荡器 (DCO) 电路，每个 DCO 功耗为 320.4μW，最大吞吐量为 7.46 亿像素/秒。
A fully scalable light-weight integrate-and-fire neuromorphic core with brain-inspired leak and local lateral inhibition features is implemented in 65nm. The core computes the neural net algorithm entirely in the time domain using standard digital circuits. A parallel two-layer architecture realized using the proposed core achieves a 91% handwritten digit recognition accuracy. The 0.24mm2 neuromorphic core including 64 digitally controlled oscillator (DCO) circuits consumes 320.4μW per DCO at a maximum throughput of 746M pixels/s.