Filters








22 Hits in 1.7 sec

PyDial: A Multi-domain Statistical Dialogue System Toolkit

Stefan Ultes, Lina M. Rojas Barahona, Pei-Hao Su, David Vandyke, Dongho Kim, Iñigo Casanueva, Paweł Budzianowski, Nikola Mrkšić, Tsung-Hsien Wen, Milica Gasic, Steve Young
2017 Proceedings of ACL 2017, System Demonstrations  
To alleviate this, we present PyDial, an opensource end-to-end statistical spoken dialogue system toolkit which provides implementations of statistical approaches for all dialogue system modules.  ...  Statistical Spoken Dialogue Systems have been around for many years.  ...  Hence, to stimulate research and make it easy for people to get involved in statistical spoken dialogue systems, we present PyDial, a multi-domain statistical spoken dialogue system toolkit.  ... 
doi:10.18653/v1/p17-4013 dblp:conf/acl/UltesRSVKCBMWGY17 fatcat:ycmpcpszavcebhzsg6a6i2av54

Using a Deep Learning Dialogue Research Toolkit in a Multilingual Multidomain Practical Application

Graham Wilcock
2018 Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence  
The toolkit (PyDial) supports research on deep reinforcement learning for multi-domain dialogues.  ...  The application (CityTalk) is a spoken dialogue system for robots that give information to tourists about local hotels and restaurants.  ...  Acknowledgments The PyDial toolkit was developed by Cambridge University Engineering Department. We thank the Dialogue Systems Group for making it available as open source.  ... 
doi:10.24963/ijcai.2018/869 dblp:conf/ijcai/Wilcock18 fatcat:56ahheov2ffarhml5lvz62ohvq

A Benchmarking Environment for Reinforcement Learning Based Task Oriented Dialogue Management [article]

Iñigo Casanueva, Paweł Budzianowski, Pei-Hao Su, Nikola Mrkšić, Tsung-Hsien Wen, Stefan Ultes, Lina Rojas-Barahona, Steve Young, Milica Gašić
2018 arXiv   pre-print
Both the environments and policy models are implemented using the publicly available PyDial toolkit and released on-line, in order to establish a testbed framework for further experiments and to facilitate  ...  To avoid the significant effort needed to hand-craft the required dialogue flow, the Dialogue Management (DM) module can be cast as a continuous Markov Decision Process (MDP) and trained through Reinforcement  ...  PyDial PyDial [39] is an open-source statistical spoken dialogue system toolkit which provides domainindependent implementations of all the dialogue system modules shown in Figure 1 , as well as simulated  ... 
arXiv:1711.11023v2 fatcat:u3ymmb2akfcgppp5jik3kqssna

Reward-Balancing for Statistical Spoken Dialogue Systems using Multi-objective Reinforcement Learning [article]

Stefan Ultes, Paweł Budzianowski, Iñigo Casanueva, Nikola Mrkšić, Lina Rojas-Barahona, Pei-Hao Su, Tsung-Hsien Wen, Milica Gašić, Steve Young
2017 arXiv   pre-print
We apply our proposed method to find optimized component weights for six domains and compare them to a default baseline.  ...  To render this search feasible, we use multi-objective reinforcement learning to significantly reduce the number of training dialogues required.  ...  The corresponding source code is included in the PyDial toolkit which can be found on www.pydial.org.  ... 
arXiv:1707.06299v1 fatcat:febpopz2zvhudc3jtdobt3hqgm

Reward-Balancing for Statistical Spoken Dialogue Systems using Multi-objective Reinforcement Learning

Stefan Ultes, Paweł Budzianowski, Iñigo Casanueva, Nikola Mrkšić, Lina M. Rojas Barahona, Pei-Hao Su, Tsung-Hsien Wen, Milica Gasic, Steve Young
2017 Proceedings of the 18th Annual SIGdial Meeting on Discourse and Dialogue  
We apply our proposed method to find optimized component weights for six domains and compare them to a default baseline.  ...  To render this search feasible, we use multi-objective reinforcement learning to significantly reduce the number of training dialogues required.  ...  The corresponding source code is included in the PyDial toolkit which can be found on www.pydial.org.  ... 
doi:10.18653/v1/w17-5509 dblp:conf/sigdial/UltesBCMRSWGY17 fatcat:fwkbkq5bqrcifaz24gzs3juxyi

DialPort, Gone Live: An Update After A Year of Development

Kyusong Lee, Tiancheng Zhao, Yulun Du, Edward Cai, Allen Lu, Eli Pincus, David Traum, Stefan Ultes, Lina M. Rojas Barahona, Milica Gasic, Steve Young, Maxine Eskenazi
2017 Proceedings of the 18th Annual SIGdial Meeting on Discourse and Dialogue  
At present six systems are linked to a central portal that directs the user to the applicable system and suggests systems that the user may be interested in.  ...  User data has started to flow into the system.  ...  The current database has just over 100 restaurants and is implemented using the multi-domain statistical dialogue system toolkit PyDial .  ... 
doi:10.18653/v1/w17-5521 dblp:conf/sigdial/LeeZDCLPTURGYE17 fatcat:tpxaxlwuyfcflp4uzh6z63li2a

PyOpenDial: A Python-based Domain-Independent Toolkit for Developing Spoken Dialogue Systems with Probabilistic Rules

Youngsoo Jang, Jongmin Lee, Jaeyoung Park, Kyeng-Hun Lee, Pierre Lison, Kee-Eung Kim
2019 Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP): System Demonstrations  
We present PyOpenDial, a Python-based domain-independent, open-source toolkit for spoken dialogue systems.  ...  To this end, we re-implemented OpenDial in Python and extended the toolkit with a number of novel functionalities for neural dialogue state tracking and action planning.  ...  We showed the aptitude of PyOpenDial by presenting how the negotiation dialogue domain can be implemented, seamlessly integrating with deep learning model trained for natural language generation.  ... 
doi:10.18653/v1/d19-3032 dblp:conf/emnlp/JangLPLLK19 fatcat:ntq7m6ugnzdutkal6gnrzzzxpy

Structured Hierarchical Dialogue Policy with Graph Neural Networks [article]

Zhi Chen, Xiaoyuan Liu, Lu Chen, Kai Yu
2020 arXiv   pre-print
A novel ComNet is proposed to model the structure of a hierarchical agent. The performance of ComNet is tested on composited tasks of the PyDial benchmark.  ...  Dialogue policy training for composite tasks, such as restaurant reservation in multiple places, is a practically important and challenging problem.  ...  PyDial toolkit , which supports multi-domain dialogue simulation with error models, has laid a good foundation for our composite task environment building.  ... 
arXiv:2009.10355v1 fatcat:pb44tnh2xbfrbgroyedth5rfzy

Plato Dialogue System: A Flexible Conversational AI Research Platform [article]

Alexandros Papangelis, Mahdi Namazifar, Chandra Khatri, Yi-Chia Wang, Piero Molino, Gokhan Tur
2020 arXiv   pre-print
As the field of Spoken Dialogue Systems and Conversational AI grows, so does the need for tools and environments that abstract away implementation details in order to expedite the development process,  ...  components, single- or multi-party interactions, and offline or online training of any conversational agent component.  ...  The following are some state-of-the-art and widely adopted platforms: • PyDial [11] is a toolkit for statistical modeling of multi-domain Dialogue Systems that supports Reinforcement Learning (RL) and  ... 
arXiv:2001.06463v1 fatcat:m6a2zcvjsfb2dadq4ipfdrucri

Addressing Objects and Their Relations: The Conversational Entity Dialogue Model

Stefan Ultes, Paweł Budzianowski, Iñigo Casanueva, Lina M. Rojas-Barahona, Bo-Hsiang Tseng, Yen-Chen Wu, Steve Young, Milica Gašić
2018 Proceedings of the 19th Annual SIGdial Meeting on Discourse and Dialogue  
Statistical spoken dialogue systems usually rely on a single-or multi-domain dialogue model that is restricted in its capabilities of modelling complex dialogue structures, e.g., relations.  ...  We demonstrate in a prototype implementation benefits of relation modelling on the dialogue level and show that a trained policy using these relations outperforms the multi-domain baseline.  ...  Acknowledgments This research was partly funded by the EPSRC grant EP/M018946/1 Open Domain Statistical Spoken Dialogue Systems.  ... 
doi:10.18653/v1/w18-5032 dblp:conf/sigdial/UltesBCRTWYG18 fatcat:unrbxnm54rcyzad6gjn7i6gyie

Domain Complexity and Policy Learning in Task-Oriented Dialogue Systems [chapter]

Alexandros Papangelis, Stefan Ultes, Yannis Stylianou
2018 Lecture Notes in Electrical Engineering  
In the present paper, we conduct a comparative evaluation of a multitude of information-seeking domains, using two well-known but fundamentally different algorithms for policy learning: GP-SARSA and DQN  ...  Our goal is to gain an understanding of how the nature of such domains influences performance.  ...  We trained the above algorithms with the PyDial toolkit [13] , using the following reward functions: for the GPS, we assign a turn penalty of -1 for each turn and a reward of +20 at the end of each successful  ... 
doi:10.1007/978-3-319-92108-2_8 fatcat:kpgqwlsum5hulibunx3ioputri

Addressing Objects and Their Relations: The Conversational Entity Dialogue Model [article]

Stefan Ultes, Paweł Budzianowski, Iñigo Casanueva, Lina Rojas-Barahona, Bo-Hsiang Tseng, Yen-Chen Wu, Steve Young, Milica Gašić
2019 arXiv   pre-print
Statistical spoken dialogue systems usually rely on a single- or multi-domain dialogue model that is restricted in its capabilities of modelling complex dialogue structures, e.g., relations.  ...  We demonstrate in a prototype implementation benefits of relation modelling on the dialogue level and show that a trained policy using these relations outperforms the multi-domain baseline.  ...  Acknowledgments This research was partly funded by the EPSRC grant EP/M018946/1 Open Domain Statistical Spoken Dialogue Systems.  ... 
arXiv:1901.01466v1 fatcat:75khwajupjcsjoxdr5yuvequhe

LD-SDS: Towards an Expressive Spoken Dialogue System based on Linked-Data [article]

Alexandros Papangelis, Panagiotis Papadakos, Margarita Kotti, Yannis Stylianou, Yannis Tzitzikas, Dimitris Plexousakis
2017 arXiv   pre-print
We envision a dialogue system named LD-SDS that will support advanced, expressive, and engaging user requests, over multiple, complex, rich, and open-domain data sources that will leverage the wealth of  ...  domains.  ...  rules; Olympus is a complete framework for implementing spoken dialogue systems (SDS) [4] ; PyDial [30] is a toolkit for developing and training statistical multi-domain SDS based on the slot filling  ... 
arXiv:1710.02973v1 fatcat:4wwpub2bkjatbhwh7kjafhyp5i

Uncertainty Estimates for Efficient Neural Network-based Dialogue Policy Optimisation [article]

Christopher Tegho, Paweł Budzianowski, Milica Gašić
2017 arXiv   pre-print
In statistical dialogue management, the dialogue manager learns a policy that maps a belief state to an action for the system to perform.  ...  This paper examines approaches to extract uncertainty estimates from deep Q-networks (DQN) in the context of dialogue management.  ...  Evaluation and results Experiments are conducted using the Cambridge restaurant domain from the PyDial toolkit [22] with a goal-driven user simulator on the semantic level [19] .  ... 
arXiv:1711.11486v1 fatcat:4dzdxbz4ufaxvcq7whgbu3bzsa

Survey on Evaluation Methods for Dialogue Systems [article]

Jan Deriu, Alvaro Rodrigo, Arantxa Otegi, Guillermo Echegoyen, Sophie Rosset, Eneko Agirre, Mark Cieliebak
2019 arXiv   pre-print
In this paper we survey the methods and concepts developed for the evaluation of dialogue systems. Evaluation is a crucial part during the development process.  ...  For this, we differentiate between the various classes of dialogue systems (task-oriented dialogue systems, conversational dialogue systems, and question-answering dialogue systems).  ...  This toolkit not only provides domain-independent implementations of different modules in a dialogue system, but also simulates users (see Section 3.4.2).  ... 
arXiv:1905.04071v1 fatcat:kyeepr7zpzgivhqdkjzfpskv3i
« Previous Showing results 1 — 15 out of 22 results