Effect of Non-linear Deep Architecture in Sequence Labeling

By Mengqiu Wang, Christopher D. Manning. 2013

57 citations recorded by the source corpus.

Subfields: 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 Natural Language Processing

1. Christopher D. Manning
2. Richard Socher
3. Greg Durrett
4. Mohit Bansal
5. Percy Liang
6. Luke S. Zettlemoyer
7. Yoav Goldberg
8. Daniel Klein

In Model Architectures

1. Christopher D. Manning
2. Richard Socher
3. Daniel Klein
4. Tao Ge
5. Baobao Chang
6. Jeffrey Pennington
7. Greg Durrett
8. Wenzhe Pei

Citations in the corpus

• GloVe: Global Vectors for Word Representation (2014, methodology): “Otherwise all config-urations are identical to those used by Wang and Manning (2013).”
• Neural CRF Parsing (2015, background): “Although other modeling choices are possible, these two points in the design space reflect common choices in NLP, and past work has suggested that nonlinear functions of indicators or linear functions of dense features may perform less well (Wang and Manning, 2013).”
• Named Entity Recognition in Chinese Clinical Text Using Deep Neural Network (2015, background): “[31] revealed that low dimension continuous space representation works significantly better in DNN than in the traditional systems (e.”
• Max-Margin Tensor Neural Network for Chinese Word Segmentation (2014, background): “In fact, CRF can be regarded as a special neural network without non-linear function (Wang and Manning, 2013). Wang and Manning (2013) conduct an empirical study on the effect of non-linearity and the results suggest that non-linear models are highly effective only when distributed representation is”