A solution to the learning dilemma for recurrent networks of spiking neurons
A solution to the learning dilemma for recurrent networks of spiking neurons
复制标题
DOI：
10.1038/s41467-020-17236-y
复制
发表时间：
2020-07-17
影响因子：
16.6
通讯作者：
Maass, Wolfgang
中科院分区：
文献类型：
--
作者：
Bellec, Guillaume;Scherr, Franz;Maass, Wolfgang
反复连接的尖峰神经元网络是大脑惊人的信息处理能力的基础。然而，尽管进行了广泛的研究，他们如何通过突触的可塑性学习来进行复杂的网络计算仍然不清楚。我们认为，这个谜题中的两个部分是由神经科学的实验数据提供的。一个数学结果告诉我们，需要如何将这些部分结合在一起，才能通过梯度下降，特别是深度强化学习，实现生物学上可信的在线网络学习。这种名为e-prop的学习方法接近时间反向传播(BPTT)的性能，BPTT是机器学习中训练递归神经网络的最著名方法。此外，它还提出了一种在基于节能尖峰的人工智能硬件中进行强大的芯片学习的方法。
Recurrently connected networks of spiking neurons underlie the astounding information processing capabilities of the brain. Yet in spite of extensive research, how they can learn through synaptic plasticity to carry out complex network computations remains unclear. We argue that two pieces of this puzzle were provided by experimental data from neuroscience. A mathematical result tells us how these pieces need to be combined to enable biologically plausible online network learning through gradient descent, in particular deep reinforcement learning. This learning method-called e-prop-approaches the performance of backpropagation through time (BPTT), the best-known method for training recurrent neural networks in machine learning. In addition, it suggests a method for powerful on-chip learning in energy-efficient spike-based hardware for artificial intelligence.