Contrastive Hebbian Feedforward Learning for Neural Networks
Contrastive Hebbian Feedforward Learning for Neural Networks
复制标题
DOI：
10.1109/tnnls.2019.2927957
复制
发表时间：
2020-06
影响因子：
10.4
通讯作者：
N. Kermiche
中科院分区：
文献类型：
--
作者：
N. Kermiche
本文讨论了在Boltzmann机器中使用的反向传播(BP)和对比Hebbian学习(CHL)的生物学合理性。本文的主要观点是，CHL是一种通用的学习算法，可以用来引导前馈网络朝向期望的结果，并引导它们远离不良结果，而不需要BP的专门反馈电路或Boltzmann机器使用的对称连接。在学习阶段向网络中的所有神经元添加扰动后，基于网络预测将多个前馈结果分类为Hebbian集和反Hebbian集。该算法适用于优化BP优胜的损失目标的网络，也适用于BP不易应用的具有随机二进制输出的网络。该算法的强大之处在于它的简单性，通过随机二进制激活将学习和梯度估计结合到单个局部Hebbian规则中。我们还将展示，Hebbian和反Hebbian关联都是从现成的信号中评估的，这些信号与Boltzmann机器中使用的CHL根本不同。我们将证明，新的学习范式，其中Hebbian/反Hebbian关联基于正确/错误的预测，是一个强大的概念，将本论文与其他生物启发的学习算法区分开来。
This paper addresses the biological plausibility of both backpropagation (BP) and contrastive Hebbian learning (CHL) used in the Boltzmann machines. The main claim of this paper is that CHL is a general learning algorithm that can be used to steer feedforward networks toward desirable outcomes, and steer them away from undesirable outcomes without any need for the specialized feedback circuit of BP or the symmetric connections used by the Boltzmann machines. After adding perturbations during the learning phase to all the neurons in the network, multiple feedforward outcomes are classified into Hebbian and anti-Hebbian sets based on the network predictions. The algorithm is applied to networks when optimizing a loss objective where BP excels and is also applied to networks with stochastic binary outputs where BP cannot be easily applied. The power of the proposed algorithm lies in its simplicity where both learning and gradient estimation through stochastic binary activations are combined into a single local Hebbian rule. We will also show that both Hebbian and anti-Hebbian correlations are evaluated from the readily available signals that are fundamentally different from CHL used in the Boltzmann machines. We will demonstrate that the new learning paradigm where Hebbian/anti-Hebbian correlations are based on correct/incorrect predictions is a powerful concept that separates this paper from other biologically inspired learning algorithms.