Brain-Inspired Learning on Neuromorphic Substrates
Brain-Inspired Learning on Neuromorphic Substrates
复制标题
DOI：
10.1109/jproc.2020.3045625
复制
发表时间：
2020-10
影响因子：
20.6
通讯作者：
Friedemann Zenke;E. Neftci
中科院分区：
文献类型：
--
作者：
Friedemann Zenke;E. Neftci
神经形态硬件致力于模拟类脑神经网络，因此有望在时间数据流上进行可扩展的低功耗信息处理。然而，要解决现实世界的问题，这些网络需要训练。然而，由于离线特性和基于梯度的学习算法所需的非局部计算，在神经形态基底上的训练产生了重大挑战。本文提供了一个数学框架的设计实用的在线学习算法的神经形态基板。具体来说，我们展示了实时递归学习（RTRL），一种用于计算传统递归神经网络（RNN）中梯度的在线算法，与用于训练尖峰神经网络（SNN）的生物学上合理的学习规则之间的直接联系。此外，我们激励基于块对角雅可比矩阵的稀疏近似，这降低了算法的计算复杂度，减少了非局部信息的要求，并根据经验导致良好的学习性能，从而提高其适用性神经形态基板。总之，我们的框架弥合了突触可塑性和深度学习中基于梯度的方法之间的差距，并为未来神经形态硬件系统上的强大信息处理奠定了基础。
Neuromorphic hardware strives to emulate brain-like neural networks and thus holds the promise for scalable, low-power information processing on temporal data streams. Yet, to solve real-world problems, these networks need to be trained. However, training on neuromorphic substrates creates significant challenges due to the offline character and the required nonlocal computations of gradient-based learning algorithms. This article provides a mathematical framework for the design of practical online learning algorithms for neuromorphic substrates. Specifically, we show a direct connection between real-time recurrent learning (RTRL), an online algorithm for computing gradients in conventional recurrent neural networks (RNNs), and biologically plausible learning rules for training spiking neural networks (SNNs). Furthermore, we motivate a sparse approximation based on block-diagonal Jacobians, which reduces the algorithm’s computational complexity, diminishes the nonlocal information requirements, and empirically leads to good learning performance, thereby improving its applicability to neuromorphic substrates. In summary, our framework bridges the gap between synaptic plasticity and gradient-based approaches from deep learning and lays the foundations for powerful information processing on future neuromorphic hardware systems.