We adapt the greedy stack LSTM dependency parser of Dyer et al. (2015) to support a training-with-exploration procedure using dynamic oracles (Goldberg and Nivre, 2013) instead of assuming an error-free action history. This form of training, which accounts for model predictions at training time, improves parsing accuracies. We discuss some modifications needed in order to get training with exploration to work well for a probabilistic neural network dependency parser.
|Title of host publication||EMNLP 2016 - Conference on Empirical Methods in Natural Language Processing, Proceedings|
|Publisher||Association for Computational Linguistics (ACL)|
|Number of pages||6|
|State||Published - 2016|
|Event||2016 Conference on Empirical Methods in Natural Language Processing, EMNLP 2016 - Austin, United States|
Duration: 1 Nov 2016 → 5 Nov 2016
|Name||EMNLP 2016 - Conference on Empirical Methods in Natural Language Processing, Proceedings|
|Conference||2016 Conference on Empirical Methods in Natural Language Processing, EMNLP 2016|
|Period||1/11/16 → 5/11/16|
Bibliographical noteFunding Information:
This work was sponsored in part by the U. S. Army Research Laboratory and the U. S. Army Research Office under contract/grant number W911NF-10-1-0533, and in part by NSF CAREER grant IIS-1054319. Miguel Ballesteros was supported by the European Commission under the contract numbers FP7-ICT-610411 (project MULTISENSOR) and H2020-RIA-645012 (project KRISTINA). Yoav Goldberg is supported by the Intel Collaborative Research Institute for Computational Intelligence (ICRI-CI), a Google Research Award and the Israeli Science Foundation (grant number 1555/15).
© 2016 Association for Computational Linguistics