Haoyuan Peng

Character-based parsing with convolutional neural network

2015 · International Conference on Artificial Intelligence

We describe a novel convolutional neural network architecture with k-max pooling layer that is able to successfully recover the structure of Chinese sentences. This network can capture active features for unseen segments of a sentence …

Learning Context-Specific Word/Character Embeddings

2017 · Proceedings of the AAAI Conference on Artificial Intelligence

Unsupervised word representations have demonstrated improvements in predictive generalization on various NLP tasks. Most of the existing models are in fact good at capturing the relatedness among words rather than their ''genuine'' similarity because the …