Injecting Designers' Knowledge in Conversational Neural Network Systems

Sequence-to-sequence neural networks are redesigning dialog managers for Conversational AI in industries. However, industrial applications impose two important constraints: training data are often scarce and the behavior of dialog managers should be strictly controlled and certified. In this paper, we propose the Conversational Logic Injected Neural Network (CLINN). This novel network merges dialog managers “programmed” using logical rules and a Sequenceto-Sequence Neural Network. We experimented with the Restaurant topic of the MultiWOZ dataset. Results show that injected rules are effective when training data set are scarce as well as when more data are available.

[1]  Andrea Favalli,et al.  Iride: an Industrial Perspective on Production Grade End to End Dialog System , 2019, CLiC-it.

[2]  Jason Weston,et al.  Wizard of Wikipedia: Knowledge-Powered Conversational agents , 2018, ICLR.

[3]  Gökhan Tür,et al.  Building a Conversational Agent Overnight with Dialogue Self-Play , 2018, ArXiv.

[4]  Geoffrey Zweig,et al.  Hybrid Code Networks: practical and efficient end-to-end dialog control with supervised and reinforcement learning , 2017, ACL.

[5]  Min-Yen Kan,et al.  Sequicity: Simplifying Task-oriented Dialogue Systems with Single Sequence-to-Sequence Architectures , 2018, ACL.

[6]  Zhijian Ou,et al.  Task-Oriented Dialog Systems that Consider Multiple Appropriate Responses under the Same Context , 2019, AAAI.

[7]  Iñigo Casanueva,et al.  Data Collection and End-to-End Learning for Conversational AI , 2019, EMNLP/IJCNLP.

[8]  Matthew Henderson,et al.  Training Neural Response Selection for Task-Oriented Dialogue Systems , 2019, ACL.

[9]  Jakob Grue Simonsen,et al.  A Hierarchical Recurrent Encoder-Decoder for Generative Context-Aware Query Suggestion , 2015, CIKM.

[10]  Fabio Massimo Zanzotto.  Human-in-the-loop Artificial Intelligence , 2017, ArXiv.

[11]  Pierre Lison,et al.  OpenDial: A Toolkit for Developing Spoken Dialogue Systems with Probabilistic Rules , 2016, ACL.

[12]  Joelle Pineau,et al.  Building End-To-End Dialogue Systems Using Generative Hierarchical Neural Network Models , 2015, AAAI.

[13]  Stefan Ultes,et al.  MultiWOZ - A Large-Scale Multi-Domain Wizard-of-Oz Dataset for Task-Oriented Dialogue Modelling , 2018, EMNLP.