Multi-domain Dialog State Tracking using Recurrent Neural Networks

Mrkšić, Nikola; Séaghdha, Diarmuid Ó; Thomson, Blaise; Gašić, Milica; Su, Pei-Hao; Vandyke, David; Wen, Tao; Young, Steve

doi:10.3115/v1/p15-2130

Cited by 151 publications

(80 citation statements)

References 11 publications

(19 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2.1.2 Hard and Soft Parameter Sharing. Most MTL approaches share the same base structure for feature extraction [1,9,12,18,26,27,33,[51][52][53] and then continue to branch out, intertwine or widen the model's parameter space. Sharing is an essential part of MTL and can be categorized as hard sharing or soft sharing.…”

Section: 11mentioning

confidence: 99%

Learning Task Relatedness in Multi-Task Learning for Images in Context

Strezoski

Noord

Worring

2019

Proceedings of the 2019 on International Conference on Multimedia Retrieval

View full text Add to dashboard Cite

Multimedia applications often require concurrent solutions to multiple tasks. These tasks hold clues to each-others solutions, however as these relations can be complex this remains a rarely utilized property. When task relations are explicitly defined based on domain knowledge multi-task learning (MTL) offers such concurrent solutions, while exploiting relatedness between multiple tasks performed over the same dataset. In most cases however, this relatedness is not explicitly defined and the domain expert knowledge that defines it is not available. To address this issue, we introduce Selective Sharing, a method that learns the inter-task relatedness from secondary latent features while the model trains. Using this insight, we can automatically group tasks and allow them to share knowledge in a mutually beneficial way. We support our method with experiments on 5 datasets in classification, regression, and ranking tasks and compare to strong baselines and state-of-the-art approaches showing a consistent improvement in terms of accuracy and parameter counts. In addition, we perform an activation region analysis showing how Selective Sharing affects the learned representation.

show abstract

Section: 11mentioning

confidence: 99%

Learning Task Relatedness in Multi-Task Learning for Images in Context

Strezoski

Noord

Worring

2019

Proceedings of the 2019 on International Conference on Multimedia Retrieval

View full text Add to dashboard Cite

show abstract

“…In machine learning in general much research has looked at adaptation of statistical models [21,22,23] however research into adaptation of SDS components to new domains [24,25,26,27,28] or user behaviour [29] presents its own challenges and is comparatively nascent. Research into these questions is growing though [30], and will continue to given the natural progression towards multi-domain SDS [31,32,33].…”

Section: Related Workmentioning

confidence: 99%

Multi-domain dialogue success classifiers for policy training

Vandyke

Gašić

et al. 2015

2015 IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU)

Self Cite

View full text Add to dashboard Cite

We propose a method for constructing dialogue success classifiers that are capable of making accurate predictions in domains unseen during training. Pooling and adaptation are also investigated for constructing multi-domain models when data is available in the new domain. This is achieved by reformulating the features input to the recurrent neural network models introduced in [1]. Importantly, on our task of main interest, this enables policy training in a new domain without the dialogue success classifier (which forms the reinforcement learning reward function) ever having seen data from that domain before. This occurs whilst incurring only a small reduction in performance relative to developing and using an in-domain dialogue success classifier. Finally, given the motivation with these dialogue success classifiers is to enable policy training with real users, we demonstrate that these initial policy training results obtained with a simulated user carry over to learning from paid human users.Index Terms-statistical spoken dialogue systems, dialogue success, multi-domain, policy training

show abstract

“…These models depend on delexicalization, using generic tags to replace specific slot types and values, and handcrafted semantic dictionaries. In practice, it is difficult to scale these models for every slot type and recent state-of-the-art models for DST use deep learning based methods to learn general representations for user and system utterances and previous system actions, and predict the turn state (Henderson et al, 2013(Henderson et al, , 2014bMrkšić et al, 2015Hori et al, 2016;Liu and Lane, 2017;Dernoncourt et al, 2017;Chen et al, 2016). However, these systems are found to perform poorly on rare and unknown slot-value pairs which was recently addressed through local slot-specific encoders (Zhong et al, 2018) and pointer network (Xu and Hu, 2018).…”

Section: Related Workmentioning

confidence: 99%

Improving Dialogue State Tracking by Discerning the Relevant Context

Sharma

Choubey

Huang

2019

Proceedings of the 2019 Conference of the North

View full text Add to dashboard Cite

A typical conversation comprises of multiple turns between participants where they go back-and-forth between different topics. At each user turn, dialogue state tracking (DST) aims to estimate user's goal by processing the current utterance. However, in many turns, users implicitly refer to the previous goal, necessitating the use of relevant dialogue history. Nonetheless, distinguishing relevant history is challenging and a popular method of using dialogue recency for that is inefficient. We, therefore, propose a novel framework for DST that identifies relevant historical context by referring to the past utterances where a particular slot-value changes and uses that together with weighted system utterance to identify the relevant context. Specifically, we use the current user utterance and the most recent system utterance to determine the relevance of a system utterance. Empirical analyses show that our method improves joint goal accuracy by 2.75% and 2.36% on WoZ 2.0 and Mul-tiWoZ 2.0 restaurant domain datasets respectively over the previous state-of-the-art GLAD model.

show abstract

Multi-domain Dialog State Tracking using Recurrent Neural Networks

Cited by 151 publications

References 11 publications

Learning Task Relatedness in Multi-Task Learning for Images in Context

Learning Task Relatedness in Multi-Task Learning for Images in Context

Multi-domain dialogue success classifiers for policy training

Improving Dialogue State Tracking by Discerning the Relevant Context

Contact Info

Product

Resources

About