Multi-turn Dialogue Response Generation in an Adversarial Learning Framework

Oluwatobi Olabiyi, Alan O Salimov, Anish Khazane, Erik Mueller

Abstract
We propose an adversarial learning approach for generating multi-turn dialogue responses. Our proposed framework, hredGAN, is based on conditional generative adversarial networks (GANs). The GAN’s generator is a modified hierarchical recurrent encoder-decoder network (HRED) and the discriminator is a word-level bidirectional RNN that shares context and word embeddings with the generator. During inference, noise samples conditioned on the dialogue history are used to perturb the generator’s latent space to generate several possible responses. The final response is the one ranked best by the discriminator. The hredGAN shows improved performance over existing methods: (1) it generalizes better than networks trained using only the log-likelihood criterion, and (2) it generates longer, more informative and more diverse responses with high utterance and topic relevance even with limited training data. This performance improvement is demonstrated on the Movie triples and Ubuntu dialogue datasets with both the automatic and human evaluations.
Anthology ID:
W19-4114
Volume:
Proceedings of the First Workshop on NLP for Conversational AI
Month:
August
Year:
2019
Address:
Florence, Italy
Editors:
Yun-Nung Chen, Tania Bedrax-Weiss, Dilek Hakkani-Tur, Anuj Kumar, Mike Lewis, Thang-Minh Luong, Pei-Hao Su, Tsung-Hsien Wen
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
121–132
Language:
URL:
https://aclanthology.org/W19-4114
DOI:
10.18653/v1/W19-4114
Bibkey:
Cite (ACL):
Oluwatobi Olabiyi, Alan O Salimov, Anish Khazane, and Erik Mueller. 2019. Multi-turn Dialogue Response Generation in an Adversarial Learning Framework. In Proceedings of the First Workshop on NLP for Conversational AI, pages 121–132, Florence, Italy. Association for Computational Linguistics.
Cite (Informal):
Multi-turn Dialogue Response Generation in an Adversarial Learning Framework (Olabiyi et al., ACL 2019)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingest-acl-2023-videos/W19-4114.pdf
PDF Search