arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1705.09189cs.CL
收藏

基于无监督Tree-LSTM的句子嵌入和句子重构联合学习

Jointly Learning Sentence Embeddings and Syntax with Unsupervised Tree-LSTMs

University of Cambridge(剑桥大学);DeepMind(深度思维)

Jean Maillard, Stephen Clark, Dani Yogatama

摘要:

我们介绍了一个神经网络,表示句子组成的话,根据诱导二叉分析树。我们使用Tree-LSTM作为我们的组合函数,沿着完全可微的自然语言图表解析器发现的树结构应用。我们的模型同时优化了组合函数和解析器,从而消除了Tree-LSTM通常需要的外部提供的解析树。因此,它可以被看作是一个基于树的RNN,相对于解析树来说是无监督的。由于它是完全可微的,我们的模型很容易用现成的梯度下降方法和反向传播进行训练。我们证明,与各种监督Tree-LSTM架构相比,它在文本蕴涵任务和反向字典任务上实现了更好的性能。

英文摘要:

We introduce a neural network that represents sentences by composing their words according to induced binary parse trees. We use Tree-LSTM as our composition function, applied along a tree structure found by a fully differentiable natural language chart parser. Our model simultaneously optimises both the composition function and the parser, thus eliminating the need for externally-provided parse trees which are normally required for Tree-LSTM. It can therefore be seen as a tree-based RNN that is unsupervised with respect to the parse trees. As it is fully differentiable, our model is easily trained with an off-the-shelf gradient descent method and backpropagation. We demonstrate that it achieves better performance compared to various supervised Tree-LSTM architectures on a textual entailment task and a reverse dictionary task.

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。

↑