A 0.086-mm2 12.7-pJ/SOP 64k-Synapse 256-Neuron Online-Learning Digital Spiking Neuromorphic Processor in 28-nm CMOS
A 0.086-mm2 12.7-pJ/SOP 64k-Synapse 256-Neuron Online-Learning Digital Spiking Neuromorphic Processor in 28-nm CMOS
复制标题
DOI：
10.1109/tbcas.2018.2880425
复制
发表时间：
2019-02-01
影响因子：
5.1
通讯作者：
Bol, David
中科院分区：
文献类型：
--
作者：
Frenkel, Charlotte;Lefebvre, Martin;Bol, David
将计算架构从冯·诺依曼转移到基于事件的尖峰神经网络（SNN），揭示了在视觉或感觉运动控制等应用中低功耗处理感觉数据的新机会。探索认知SNN的道路需要设计紧凑，低功耗和多功能的实验平台，其关键要求是在线学习，以便在不受控制的环境中适应和学习新功能。然而，在SNN中嵌入在线学习目前受到高复杂性和区域开销的阻碍。在本文中，我们提出了ODIN，一个0.086毫米（2）64 k突触256神经元在线学习数字尖峰神经形态处理器在28纳米FDSOI CMOS实现的最小能量每突触操作（SOP）的12.7 pJ。它利用尖峰驱动的突触可塑性（SDSP）的学习规则的高密度嵌入式在线学习的有效实现，每个4位突触只有0.68 μ m（2）。神经元可以独立地配置为标准的泄漏积分和发射模型，或作为自定义的现象学模型，模拟在生物尖峰神经元中发现的20个Izhikevich行为。通过将6 k 16 x 16 MNIST训练图像单一呈现给具有基于片上SDS学习的单层全连接10神经元网络，ODIN实现了84.5%的分类准确率，同时使用秩序编码在0.55 V时仅消耗15 nJ/推理。因此，ODIN使认知神经形态设备的进一步发展，低功耗，自适应和低成本的处理。
Shifting computing architectures from von Neumann to event-based spiking neural networks (SNNs) uncovers new opportunities for low-power processing of sensory data in applications such as vision or sensorimotor control. Exploring roads toward cognitive SNNs requires the design of compact, low-power and versatile experimentation platforms with the key requirement of online learning in order to adapt and learn new features in uncontrolled environments. However, embedding online learning in SNNs is currently hindered by high incurred complexity and area overheads. In this paper, we present ODIN, a 0.086-mm(2) 64k-synapse 256-neuron online-learning digital spiking neuromorphic processor in 28-nm FDSOI CMOS achieving a minimum energy per synaptic operation (SOP) of 12.7 pJ. It leverages an efficient implementation of the spike-driven synaptic plasticity (SDSP) learning rule for high-density embedded online learning with only 0.68 mu m(2) per 4-bit synapse. Neurons can be independently configured as a standard leaky integrate-and-fire model or as a custom phenomenological model that emulates the 20 Izhikevich behaviors found in biological spiking neurons. Using a single presentation of 6k 16 x 16 MNIST training images to a single-layer fully-connected 10-neuron network with on-chip SDSP-based learning, ODIN achieves a classification accuracy of 84.5%, while consuming only 15 nJ/inference at 0.55 V using rank order coding. ODIN thus enables further developments toward cognitive neuromorphic devices for low-power, adaptive and low-cost processing.