Research data supporting "On-line Active Reward Learning for Policy Optimisation in Spoken Dialogue Systems"
University of Cambridge
MetadataShow full item record
Su, P., Gasic, M., Mrksic, N., Rojas-Barahona, L., Ultes, S., Vandyke, D., Wen, T., & et al. (2016). Research data supporting "On-line Active Reward Learning for Policy Optimisation in Spoken Dialogue Systems" [Dataset]. https://www.repository.cam.ac.uk/handle/1810/256020
This repository contains the data presented in the paper "On-line Active Reward Learning for Policy Optimisation in Spoken Dialogue Systems" in ACL 2016. Two separate datasets as described in section 4 of the paper are presented: 1. DialogueEmbedding/ It contains the [train|valid|test] data for the unsupervised dialogue embedding creation, each with *.[feature|reward|turn|subjsuc]. Note that *.turn includes the lines to be read for each dialogue in *.[feature|reward|subjsuc], and *.subjsuc is the user's subjective rating. The feature size is 74. 2. DialoguePolicy/ It includes four contrasting systems with different reward models: [GP|RNN|ObjSubj|Subj]. Inside each system directory is the data obtained in interaction with Amazon Mechanical Turk users while training three policies with same config: policy_[1|2|3]. and a .csv for the evaluation result along with the trainig process. In each policy_[1|2|3]/ there is a list of calls with a time stamp in the name which contains session.xml file for dialogue log and feedback.xml file for user feedback
This research data supports "On-line Active Reward Learning for Policy Optimisation in Spoken Dialogue Systems" which has been published in "Proceedings of Association for Computational Linguistics (ACL)".
csv, xml, README
spoken dialogue systems, deep learning, reward modelling, reinforcement learning, Gaussian process
Publication Reference: https://arxiv.org/pdf/1605.07669v2.pdf
This work was supported by the EPSRC [grant number Cambridge Trust].
This record's URL: https://www.repository.cam.ac.uk/handle/1810/256020
Attribution 2.0 UK: England & Wales
Licence URL: http://creativecommons.org/licenses/by/2.0/uk/
Recommended or similar items
The following licence files are associated with this item: