Sentence Modeling with Gated Recursive Neural Network

By Xinchi Chen, Xipeng Qiu, Chenxi Zhu, Shiyu Wu, Xuanjing Huang. 2015

32 citations recorded by the source corpus.

Recently, neural network based sentence modeling methods have achieved great progress. Among these methods, the recursive neural networks (RecNNs) can effectively model the combination of the words in sentence. However, RecNNs need a given external topological structure, like syntactic tree. In this paper, we propose a gated recursive neural network (GRNN) to model sentences, which employs a full binary tree (FBT) structure to control the combinations in recursive structure. By introducing two kinds of gates, our model can better model the complicated combinations of features. Experiments on three text classification datasets show the effectiveness of our model.

Subfields: Artificial Intelligence, Natural Language Processing, Model Architectures

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Gilles Louppe
2. Kyle Cranmer
3. Kang Min Yoo
4. Sang-goo Lee
5. Jihun Choi
6. Samuel R. Bowman
7. C. Becot
8. Mohit Bansal

In Natural Language Processing

1. Kang Min Yoo
2. Sang-goo Lee
3. Jihun Choi
4. Gilles Louppe
5. Samuel R. Bowman
6. Kyle Cranmer
7. Kyunghyun Cho
8. Liqiang Xiao

In Model Architectures

1. Kyle Cranmer
2. Sang-goo Lee
3. Kang Min Yoo
4. Jihun Choi
5. Gilles Louppe
6. Kyunghyun Cho
7. Liqiang Xiao
8. Yongkun Wang

Citations in the corpus

• Analysis Methods in Neural Language Processing: A Survey (2018, neutral): “, 2018b) and sentence classification (Chen et al., 2015).”
• Learning to Compose Task-Specific Tree Structures (2017, methodology): “Following (Cho et al., 2014), where grConv is used as an encoder for machine translation, grConv and its variants are also applied in sentence classification tasks (Chen et al., 2015; Zhao et al., 2015).”
• Multi-Task Label Embedding for Text Classification (2017, methodology): “…focus the idea and effects of integrating label embedding with multi-task learning, so we just apply (Graves 2013) to realize Le I and Le L , which can be further implemented by other more effective sentence learning models (Liu et al. 2015a; Chen et al. 2015) and produce better performances.”
• Lattice-Based Recurrent Neural Network Encoders for Neural Machine Translation (2016, unknown): “quence models are RNNs (Mikolov et al., 2010) with LSTM (Hochreiter and Schmidhuber, 1997; Sutskever et al., 2014; Le and Zuidema, 2015a; Zhu et al., 2015; Liu et al., 2015) or GRU (Cho et al., 2014; Chen et al., 2015). Further, some researchers extend standard RNNs to non-sequential ones, such as m”
• QCD-aware recursive neural networks for jet physics (2017, background): “[14] Xinchi Chen, Xipeng Qiu, Chenxi Zhu, Shiyu Wu, and Xuanjing Huang.”