BriefGPT.xyz
Jan, 2016
循环记忆网络语言建模
Recurrent Memory Network for Language Modeling
HTML
PDF
Ke Tran, Arianna Bisazza, Christof Monz
TL;DR
本文提出了一种新的循环神经网络体系结构Recurrent Memory Network(RMN),不仅能够放大循环神经网络的作用,而且有助于我们理解其内部功能并发现数据中的潜在模式。在语言建模和句子完成任务上展示了RMN的强大性能。在长句完成挑战中,RMN的准确性为69.2%,超过了以前的最新技术水平。
Abstract
recurrent neural networks
(RNN) have obtained excellent result in many
natural language processing
(NLP) tasks. However, understanding and interpreting the source of this success remains a challenge. In this pape
→