一种可逆的深度残差神经网络结构
Reversible Architectures for Arbitrarily Deep Residual Neural Networks
摘要:
近年来,深度残差网络已成功应用于许多计算机视觉和自然语言处理任务,以更深更广的架构推动了最先进的性能。在这项工作中,我们将深层剩余网络解释为常微分方程(ODE),这在数学和物理学中已经研究了很长时间,并取得了丰富的理论和经验成功。从这个解释出发,我们发展了一个关于深度神经网络的稳定性和可逆性的理论框架,并推导出三个可逆的神经网络架构,它们可以在理论上任意深入。可逆性属性允许内存高效的实现,它不需要存储大多数隐藏层的激活。再加上我们架构的稳定性,这使得只使用适度的计算资源就可以训练更深层次的网络。我们提供了理论分析和实证结果。实验结果表明,我们的架构对CIFAR-10,CIFAR-100和STL-10的几个强大的基线具有优越或同等的最先进的性能的功效。此外,我们表明,我们的架构产生优越的结果时,使用更少的训练数据训练。
英文摘要:
Recently, deep residual networks have been successfully applied in many computer vision and natural language processing tasks, pushing the state-of-the-art performance with deeper and wider architectures. In this work, we interpret deep residual networks as ordinary differential equations (ODEs), which have long been studied in mathematics and physics with rich theoretical and empirical success. From this interpretation, we develop a theoretical framework on stability and reversibility of deep neural networks, and derive three reversible neural network architectures that can go arbitrarily deep in theory. The reversibility property allows a memory-efficient implementation, which does not need to store the activations for most hidden layers. Together with the stability of our architectures, this enables training deeper networks using only modest computational resources. We provide both theoretical analyses and empirical results. Experimental results demonstrate the efficacy of our architectures against several strong baselines on CIFAR-10, CIFAR-100 and STL-10 with superior or on-par state-of-the-art performance. Furthermore, we show our architectures yield superior results when trained using fewer training data.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。