Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Learning to skim text, Yu+, ACL'17 #82

Open
AkihikoWatanabe opened this issue Dec 31, 2017 · 2 comments
Open

Learning to skim text, Yu+, ACL'17 #82

AkihikoWatanabe opened this issue Dec 31, 2017 · 2 comments
Labels
Efficiency/SpeedUp Neural NLP

Comments

@AkihikoWatanabe
Copy link
Owner

http://aclweb.org/anthology/P/P17/P17-1172.pdf
@AkihikoWatanabe
Copy link
Owner Author

解説スライド：http://www.lr.pi.titech.ac.jp/~haseshun/acl2017suzukake/slides/07.pdf

@AkihikoWatanabe
Copy link
Owner Author

AkihikoWatanabe commented Dec 31, 2017 •
edited

image
image

いくつ読み飛ばすかも潜在変数として一緒に学習する。潜在変数（離散変数）なので、�普通に尤度最大化するやり方では学習できず、おまけに離散変数なのでバックプロパゲーション使えないので、強化学習で学習する。

Vanilla LSTMと比較し、色々なタスクで実験した結果、性能も（少し）上がるし、スピードアップもする。

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
Efficiency/SpeedUp Neural NLP
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@AkihikoWatanabe