On-line Learning of Wide-domain Spoken Dialogue Systems thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

On-line Learning of Wide-domain Spoken Dialogue Systems

Published on Jul 31, 20161143 Views

The first part of the talk reviews the general structure of limited domain statistical SDS and then explains how a collection of limited domain systems can be merged using the framework of Bayesian

Related categories

Chapter list

On - line Learning of Wide - Domain Statistical Spoken Dialogue Systems00:00
Objectives01:36
Reinforcement Learning - 103:03
Reinforcement Learning - 203:38
Single Limited Domain Statistical SDS04:12
Extending to Wide Domains10:17
Multi-domain SDS11:03
Bayesian Committee Machine13:09
Bayesian Committee Machine Training Performance13:36
Training with Real Users16:09
On - line Reward Estimation17:37
On - line Reward and Policy Learning20:33
Conclusions21:04