Learning Arithmetic Operations With A Multistep Deep Learning
Learning Arithmetic Operations With A Multistep Deep Learning
复制标题
通过多步深度学习学习算术运算
DOI：
10.1109/ijcnn48605.2020.9206963
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Frederic Armetta
中科院分区：
文献类型：
--
作者：
Bastien Nollet;Mathieu Lefort;Frederic Armetta
当深度神经网络应用于可以表示为算法程序的任务时，很难训练。在本文中，我们提出研究如何通过使用外部记忆和主动选择输入，在算法的所有步骤中对网络进行显式引导，以提高其学习能力。这个想法是从孩子的学习中获得灵感，并通过与外部支持（如纸张）的互动来完成一个过程。我们表明，将这种机制应用于简单的多层感知器，可以显著提高其在学习多位数加法或乘法时的性能，这是简单但具有挑战性的操作。
Deep neural networks are difficult to train when applied to tasks that can be expressed as algorithmic procedures. In this article, we propose to study how the explicit guidance of a network through all steps of the algorithm, using external memory and active choice of inputs, can improve its learning capability. The idea is to take inspiration from a child’s learning and running through a procedure via interaction with an external support such as a paper. We show that this mechanism applied to a simple multilayer perceptron can significantly improve its performance when learning either a multi-digit addition or multiplication, which are simple but yet challenging operations to learn.