Dual Supervised Learning for Natural Language Understanding and Generation

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 5472–5477 Florence, Italy, July 28 - August 2, 2019. c⃝2019 Association for Computational Linguistics 5472 Dual Supervised Learning for Natural Language Understanding and Generation Shang-Yu Su Chao-Wei Huang Yun-Nung Chen Department of Computer Science and Information Engineering National Taiwan University {f05921117,r07922069}@ntu.edu.tw y.v.chen@ieee.org Abstract Natural language understanding (NLU) and natural language generation (NLG) are both critical research topics in the NLP and dia- logue fields. Natural language understanding is to extract the core semantic meaning from the given utterances, while natural language generation is opposite, of which the goal is to construct corresponding sentences based on the given semantics. However, such dual re- lationship has not been investigated in liter- ature. This paper proposes a novel learning framework for natural language understand- ing and generation on top of dual supervised learning, providing a way to exploit the dual- ity. The preliminary experiments show that the proposed approach boosts the performance for both tasks, demonstrating the effectiveness of the dual relationship.1 1 Introduction Spoken dialogue systems that can help users solve complex tasks such as booking a movie ticket have become an emerging research topic in artificial in- telligence and natural language processing areas. With a well-designed dialogue system as an in- telligent personal assistant, people can accomplish certain tasks more easily via natural language in- teractions. The recent advance of deep learning has inspired many applications of neural dialogue systems (Wen et al., 2017; Bordes et al., 2017; Dhingra et al., 2017; Li et al., 2017). A typical dialogue system pipeline can be divided into sev- eral parts: 1) a speech recognizer that transcribes a user’s speech input into texts, 2) a natural lan- guage understanding module (NLU) that classifies the domain and associated intents and fills slots to form a semantic frame (Chi et al., 2017; Chen et al., 2017; Zhang et al., 2018; Su et al., 2018c, 1https://github.com/MiuLab/DualSL Natural Language Understanding Natural Language Generation Natural Language McDonald’s is a cheap restaurant nearby the station. Semantic Frame RESTAURANT=“McDonald’s” PRICE=“cheap” LOCATION= “nearby the station” Figure 1: NLU and NLG emerge as a dual form. 2019), 3) a dialogue state tracker (DST) that pre- dicts the current dialogue state in the multi-turn conversations, 4) a dialogue policy that determines the system action for the next step given the cur- rent state (Peng et al., 2018; Su et al., 2018a), and 5) a natural language generator (NLG) that outputs a response given the action semantic frame (Wen et al., 2015; Su et al., 2018b; Su and Chen, 2018). Many artificial intelligence tasks come with a dual form; that is, we could directly swap the in- put and the target of a task to formulate another task. Machine translation is a classic example (Wu et al., 2016); for example, translating from En- glish to Chinese has a dual task of translating from Chinese to English; automatic speech recognition (ASR) and text-to-speech (TTS) also have struc- tural duality (Tjandra et al., 2017). …

Content truncated to 500 words