Finding Sparse Structure for Domain Specific Neural Machine Translation

Jianze Liang
Jianze Liang
Mingxuan Wang
Mingxuan Wang
Cited by: 0|Views22
Weibo:
We propose PRUNE-TUNE, an effective way for adapting neural machine translation models which first generates an informative subnetwork for the general domain via gradual pruning and fine-tunes the unnecessary parameters for the target domain

Abstract:

Fine-tuning is a major approach for domain adaptation in Neural Machine Translation (NMT). However, unconstrained fine-tuning requires very careful hyper-parameter tuning otherwise it is easy to fall into over-fitting on the target domain and degradation on the general domain. To mitigate it, we propose PRUNE-TUNE, a novel domain adapta...More
0
Full Text
Bibtex
Weibo
Your rating :
0

 

Tags
Comments