Multi-head Attention With Disagreement Regularization
Jian Li, Zhaopeng Tu, Baosong Yang, Michael R. Lyu, Tong Zhang . Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing 2018 – 177 citations
Multi-head attention is appealing for the ability to jointly attend to information from different representation subspaces at different positions. In this work, we introduce a disagreement regularization to explicitly encourage the diversity among multiple attention heads. Specifically, we propose three types of disagreement regularization, which respectively encourage the subspace, the attended positions, and the output representation associated with each attention head to be different from other heads. Experimental results on widely-used WMT14 English-German and WMT17 Chinese-English translation tasks demonstrate the effectiveness and universality of the proposed approach.