arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1506.02516cs.NEcs.CLcs.LG
收藏

学习用无限记忆进行转换

Learning to Transduce with Unbounded Memory

Google DeepMind(谷歌DeepMind);Oxford University(牛津大学)

Edward Grefenstette, Karl Moritz Hermann, Mustafa Suleyman, Phil Blunsom

摘要:

最近,深度递归神经网络在自然语言转换问题上已经证明了强大的结果。在本文中,我们探讨了这些模型的代表性力量,使用合成语法设计表现出类似的现象,发现在真正的转导问题,如机器翻译。这些实验使我们提出了新的基于内存的递归网络,它实现了传统数据结构(如Stacks,Dockes和DeQues)的连续可微模拟。我们表明,这些架构表现出优于深度RNN的泛化性能,并且通常能够在我们的转导实验中学习底层生成算法。

英文摘要:

Recently, strong results have been demonstrated by Deep Recurrent Neural Networks on natural language transduction problems. In this paper we explore the representational power of these models using synthetic grammars designed to exhibit phenomena similar to those found in real transduction problems such as machine translation. These experiments lead us to propose new memory-based recurrent networks that implement continuously differentiable analogues of traditional data structures such as Stacks, Queues, and DeQues. We show that these architectures exhibit superior generalisation performance to Deep RNNs and are often able to learn the underlying generating algorithms in our transduction experiments.

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。

↑