Output details
15 - General Engineering
University of Cambridge
Return to search
Output 0 of 0 in the submission
Article title
Reinforcement learning for parameter estimation in statistical spoken dialogue systems
Type
D - Journal article
Title of journal
Computer Speech and Language
Article number
-
Volume number
26
Issue number
3
First page of article
168
ISSN of journal
0885-2308
Year of publication
2012
URL
-
Number of additional authors
2
Additional information
A preliminary version of this paper received a Best Paper award at the Interspeech 2010 conference, in the area of spoken language processing. It is the first paper to demonstrate that all model parameters in a spoken dialogue system can be optimised using reinforcement learning. This has particular relevance to commercial deployment since it allows a system to be optimised even when a client insists on the inclusion of sub-optimal design features (Paek and Pieraccini, Speech Communication 50(8-9): 716-729,2008)
Interdisciplinary
-
Cross-referral requested
-
Research group
None
Proposed double-weighted
No
Double-weighted statement
-
Reserve for a double-weighted output
No
Non-English
No
English abstract
-
