Multi-Task Semantic Dependency Parsing with Policy Gradient for Learning Easy-First Strategies

Shuhei Kurita, Anders Søgaard


Abstract
In Semantic Dependency Parsing (SDP), semantic relations form directed acyclic graphs, rather than trees. We propose a new iterative predicate selection (IPS) algorithm for SDP. Our IPS algorithm combines the graph-based and transition-based parsing approaches in order to handle multiple semantic head words. We train the IPS model using a combination of multi-task learning and task-specific policy gradient training. Trained this way, IPS achieves a new state of the art on the SemEval 2015 Task 18 datasets. Furthermore, we observe that policy gradient training learns an easy-first strategy.
Anthology ID:
P19-1232
Volume:
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2019
Address:
Florence, Italy
Editors:
Anna Korhonen, David Traum, Lluís Màrquez
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2420–2430
Language:
URL:
https://aclanthology.org/P19-1232
DOI:
10.18653/v1/P19-1232
Bibkey:
Cite (ACL):
Shuhei Kurita and Anders Søgaard. 2019. Multi-Task Semantic Dependency Parsing with Policy Gradient for Learning Easy-First Strategies. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 2420–2430, Florence, Italy. Association for Computational Linguistics.
Cite (Informal):
Multi-Task Semantic Dependency Parsing with Policy Gradient for Learning Easy-First Strategies (Kurita & Søgaard, ACL 2019)
Copy Citation:
PDF:
https://aclanthology.org/P19-1232.pdf
Supplementary:
 P19-1232.Supplementary.pdf
Code
 shuheikurita/semrl