Building Interpretable Interaction Trees for Deep NLP Models

This paper proposes a method to disentangle and quantify interactions among words that are encoded inside a DNN for natural language processing. We construct a tree to encode salient interactions extracted by the DNN. Six metrics are proposed to analyze properties of interactions between constituents in a sentence. The interaction is defined based on Shapley values of words, which are considered as an unbiased estimation of word contributions to the network prediction. Our method is used to quantify word interactions encoded inside the BERT, ELMo, LSTM, CNN, and Transformer networks. Experimental results have provided a new perspective to understand these DNNs, and have demonstrated the effectiveness of our method.

Authors

• Die Zhang
• Hao Zhang vivo Mobile Communication Co., Ltd
• Huilin Zhou Shanghai Jiao Tong University
• Xiaoyi Bao School of Artificial Intelligence, University of Chinese Academy of Sciences Institute of automation, Chinese Academy of Sciences Luoyang Institute for Robot and Intelligent Equipment
• Da Huo Shanghai Jiao Tong University
• Ruizhao Chen
• Xu Cheng
• Mengyue Wu Shanghai Jiao Tong University
• Quanshi Zhang Shanghai Jiao Tong University

Keywords

No keywords are indexed for this paper.