A Network-based End-to-End Trainable Task-oriented Dialogue System

Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, pages 438–449, Valencia, Spain, April 3-7, 2017. c⃝2017 Association for Computational Linguistics A Network-based End-to-End Trainable Task-oriented Dialogue System Tsung-Hsien Wen1, David Vandyke1, Nikola Mrkši´c1, Milica Gaši´c1, Lina M. Rojas-Barahona1, Pei-Hao Su1, Stefan Ultes1, and Steve Young1 1Cambridge University Engineering Department, Trumpington Street, Cambridge, CB2 1PZ, UK {thw28,djv27,nm480,mg436,lmr46,phs26,su259,sjy11}@cam.ac.uk 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 in- volves either a large amount of handcraft- ing, or acquiring costly labelled datasets to solve a statistical learning problem for each component. In this work we intro- duce a neural network-based text-in, text- out end-to-end trainable goal-oriented di- alogue 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 sys- tems 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 help- ing them to accomplish tasks in a restaurant search domain. 1 Introduction Building a task-oriented dialogue system such as a hotel booking or a technical support service is difficult because it is application-specific and there is usually limited availability of training data. To mitigate this problem, recent machine learning ap- proaches to task-oriented dialogue system design have cast the problem as a partially observable Markov Decision Process (POMDP) (Young et al., 2013) with the aim of using reinforcement learn- ing (RL) to train dialogue policies online through interactions with real users (Gaši´c et al., 2013). However, the language understanding (Henderson et al., 2014; Yao et al., 2014) and language gener- ation (Wen et al., 2015b; Wen et al., 2016) mod- ules still rely on supervised learning and therefore need corpora to train on. Furthermore, to make RL tractable, the state and action space must be carefully designed (Young et al., 2013; Young et al., 2010), which may restrict the expressive power and learnability of the model. Also, the reward functions needed to train such models are difficult to design and hard to measure at run-time (Su et al., 2015; Su et al., 2016). At the other end of the spectrum, sequence to sequence learning (Sutskever et al., 2014) has in- spired several efforts to build end-to-end trainable, non-task-oriented conversational systems (Vinyals and Le, 2015; Shang et al., 2015; Serban et al., 2015b). This family of approaches treats dialogue as a source to target sequence transduction problem, applying an encoder network (Cho et al., 2014) to encode a user query into a distributed vector rep- resenting its semantics, which then conditions a decoder network to generate each system response. These models typically require a large amount of data to train. They allow the creation of effective chatbot type systems but they lack any capability for supporting domain specific tasks, for example, being able to interact with …

Content truncated to 500 words