Meta-Learning for Low-resource Natural Language Generation in Task-oriented Dialogue Systems

By Fei Mi, Minlie Huang, Jiyong Zhang, Boi Faltings. 2019

101 citations recorded by the source corpus.

Natural language generation (NLG) is an essential component of task-oriented dialogue systems. Despite the recent success of neural approaches for NLG, they are typically developed for particular domains with rich annotated training examples. In this paper, we study NLG in a low-resource setting to generate sentences in new scenarios with handful training examples. We formulate the problem from a meta-learning perspective, and propose a generalized optimization-based approach (Meta-NLG) based on the well-recognized model-agnostic meta-learning (MAML) algorithm. Meta-NLG defines a set of meta tasks, and directly incorporates the objective of adapting to new low-resource NLG tasks into the meta-learning optimization process. Extensive experiments are conducted on a large multi-domain dataset (MultiWoz) with diverse linguistic variations. We show that Meta-NLG significantly outperforms other training procedures in various low-resource configurations. We analyze the results, and demonstrate that Meta-NLG adapts extremely fast and well to low-resource situations.

Subfields: Artificial Intelligence

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Continual Learning for Task-oriented Dialogue System with Iterative Network Pruning, Expanding and Masking (2021, neutral): “…dialogue systems (TDSs) with sequence-to-sequence (Seq2Seq) models (Eric and Manning, 2017; Madotto et al., 2018; Gangi Reddy et al., 2019; Qin et al., 2020; Mi et al., 2019; He et al., 2020; Wang et al., 2020; Qin et al., 2021), which have taken the state-of-the-art of TDSs to a new level.”
Continual Learning for Natural Language Generation in Task-oriented Dialog Systems (2020, background): “The latter is a one-time adaptation process, and the focus is to optimize performance on a target domain transferred from source domains but without considering potential performance drop on source domains (Mi et al., 2019; Qian and Yu, 2019; Peng et al., 2020).”
Few-shot Natural Language Generation for Task-Oriented Dialog (2020, background): “, 2017; Tran and Nguyen, 2017), as well as learning objectives (Zhu et al., 2019; Zhu, 2020; Mi et al., 2019).”
Meta-Learning with Implicit Gradients (2019, background): “MAML has been shown to be as expressive as black-box approaches [14], is applicable to a broad range of settings [16, 37, 1, 18], and recovers a convergent and consistent optimization procedure [13].”
Multi-Step Model-Agnostic Meta-Learning: Convergence and Improved Algorithms (2020, background): “Many follow-up studies (Grant et al., 2018; Finn et al., 2019; Jerfel et al., 2018; Finn and Levine, 2017; Finn et al., 2018; Mi et al., 2019; Liu et al., 2019; Rothfuss et al., 2018; Foerster et al., 2018; Fallah et al., 2019; Collins et al., 2020; Fallah et al., 2020) have extended MAML from diffe”
Template Guided Text Generation for Task Oriented Dialogue (2020, background): “…transfer learning in low resource settings has also been an extensively studied problem (Tran and Le Nguyen, 2018; Chen et al., 2020; Peng et al., 2020; Mi et al., 2019), with recently released datasets like SGD (Rastogi et al., 2019) and FewShotWOZ (Peng et al., 2020) providing a good benchmark.”
Low-Resource Generation of Multi-hop Reasoning Questions (2020, background): “The low-resource problem has been studied in the tasks such as machine translation (Gu et al., 2018), pos tagging (Kann et al., 2018), word embedding (Jiang et al., 2018), text generation (Wiseman et al., 2018), and dialogue systems (Mi et al., 2019).”
Self-Supervised Meta-Learning for Few-Shot Natural Language Classification Tasks (2020, background): “Such approaches (Yu et al., 2018; Gu et al., 2018; Guo et al., 2018; Han et al., 2018; Mi et al., 2019) train to generalize to new labels of a specific task like relation classification and don’t generalize to novel tasks.”
CINS: Comprehensive Instruction for Few-shot Learning in Task-oriented Dialog Systems (2021, result): “We contend that this is a more reasonable few-shot setting compared to existing fewshot ToD studies (Mi et al. 2019; Peng et al. 2020b; Kale and Rastogi 2020a; Lin et al. 2021) that use full-size validation data.”
Few-Shot Natural Language Generation by Rewriting Templates (2020, background): “Prior work (Mi et al., 2019; Tran and Le Nguyen, 2018; Wen et al., 2016) has studied zero shot learning, domain adaptation etc.”
Exploring Auxiliary Reasoning Tasks for Task-oriented Dialog Systems with Meta Cooperative Learning (2021, methodology): “For example, Mi et al. (2019) applied model-agnostic meta-learning (MAML) (Finn, Abbeel, and Levine 2017) to solve natural language generation in task-oriented dialog systems in the low-resource scenario.”
Language Models as Few-Shot Learner for Task-Oriented Dialogue Systems (2020, background): “, 2020), NLG/ACT (Mi et al., 2019), pipelined endto-end models (Qian and Yu, 2019) and personalized dialogue systems (Madotto et al.”
Learning to Few-Shot Learn Across Diverse Natural Language Classification Tasks (2019, background): “Other approaches (Han et al., 2018; Obamuyide and Vlachos, 2019; Geng et al., 2019; Mi et al., 2019) train for a specific classification task and do not generalize beyond the training task itself.”
Evaluating Semantic Accuracy of Data-to-Text Generation with Natural Language Inference (2020, methodology): “The closest commonly used metric is the slot error rate , which is typically based on pattern matching tailored for a given dataset (Reed et al., 2018; Mi et al., 2019; Dušek et al., 2020).”