论文标题
复发性神经网络链中的快速无内存预测算法
A fast memoryless predictive algorithm in a chain of recurrent neural networks
论文作者
论文摘要
在最近的出版物（ARXIV：2007.08063V2 [CS.LG]）中，提出了一种快速预测算法（RN）。在本手稿中，我们将这种方法推广到一系列RN，并表明它可以在自然神经系统中实施。当网络被递归地用于预测值的序列时，提出的算法不需要存储原始输入序列。与标准移动/扩展窗口预测过程相比，它增加了新方法的鲁棒性。我们考虑对训练有素的网络的要求，这些要求允许实施拟议的算法并在神经科学环境中进行讨论。
In the recent publication (arxiv:2007.08063v2 [cs.LG]) a fast prediction algorithm for a single recurrent network (RN) was suggested. In this manuscript we generalize this approach to a chain of RNs and show that it can be implemented in natural neural systems. When the network is used recursively to predict sequence of values the proposed algorithm does not require to store the original input sequence. It increases robustness of the new approach compared to the standard moving/expanding window predictive procedure. We consider requirements on trained networks that allow to implement the proposed algorithm and discuss them in the neuroscience context.
