A Network-based End-to-End Trainable Task-oriented Dialogue System
Abstract
Teaching machines to accomplish tasks by conversing naturally with humans is challenging. Currently, developing task-oriented dialogue systems requires creating multiple components and typically this involves either a large amount of handcrafting, or acquiring costly labelled datasets to solve a statistical learning problem for each component. In this work we introduce a neural network-based text-in, text-out end-to-end trainable goal-oriented dialogue system along with a new way of collecting dialogue data based on a novel pipe-lined Wizard-of-Oz framework. This approach allows us to develop dialogue systems easily and without making too many assumptions about the task at hand. The results show that the model can converse with human subjects naturally whilst helping them to accomplish tasks in a restaurant search domain.
Cite
@article{arxiv.1604.04562,
title = {A Network-based End-to-End Trainable Task-oriented Dialogue System},
author = {Tsung-Hsien Wen and David Vandyke and Nikola Mrksic and Milica Gasic and Lina M. Rojas-Barahona and Pei-Hao Su and Stefan Ultes and Steve Young},
journal= {arXiv preprint arXiv:1604.04562},
year = {2017}
}
Comments
published at EACL 2017