Collaborative Regularization for Bidirectional Domain Mapping

Jen Tzung Chien, Wei Hsiang Chang

研究成果: Conference contribution同行評審

4 引文 斯高帕斯(Scopus)

摘要

Learning both domain mapping and domain knowledge is crucial for different sequence-to-sequence (seq2seq) tasks. Traditionally, seq2seq model only characterized domain mapping while the knowledge in source and target domains was ignored. To strengthen seq2seq representation, this study presents a unified transformer for bidirectional domain mapping where collaborative regularization is imposed. This regularization enforces the bidirectional mapping constraint and avoids the model from overfitting for better generalization. Importantly, the unified learning objective is optimized for collaborative learning among different modules in two domains with two learning directions. Experiments on machine translation demonstrate the merit of unified transformer by comparing with the existing methods under different tasks and settings.

原文English
主出版物標題IJCNN 2021 - International Joint Conference on Neural Networks, Proceedings
發行者Institute of Electrical and Electronics Engineers Inc.
ISBN(電子)9780738133669
DOIs
出版狀態Published - 18 7月 2021
事件2021 International Joint Conference on Neural Networks, IJCNN 2021 - Virtual, Shenzhen, 中國
持續時間: 18 7月 202122 7月 2021

出版系列

名字Proceedings of the International Joint Conference on Neural Networks
2021-July

Conference

Conference2021 International Joint Conference on Neural Networks, IJCNN 2021
國家/地區中國
城市Virtual, Shenzhen
期間18/07/2122/07/21

指紋

深入研究「Collaborative Regularization for Bidirectional Domain Mapping」主題。共同形成了獨特的指紋。

引用此