Abstract
All dialogue managers must maintain an internal state representing what has happened in the dialogue. In hand-crafted approaches the internal state is directly defined by the dialogue system designer. More recently, probabilistic approaches have been developed to better handle the uncertainty inherent in dialogue. As mentioned in the previous chapter, the probabilistic approach defines the internal state as a belief distribution over environment states.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
In Williams et al. (2005) the notation is slightly different with \(s_u = g\), \(a_u = u\) and \(s_d = h\).
- 2.
The term slot usually implies a concept which the user must specify a particular value for. In the case of a Bayesian network the concept values might be inferred from other information so the term concept will be preferred here.
- 3.
Indeed, Williams (2007a) has discussed a Bayesian network structure for the use of POMDPs in troubleshooting tasks. Nodes in the resulting network include whether there is “no power” or “no network” for an internet user. Various different variables could be used in other applications.
- 4.
The algorithm is generally known as the sum-product algorithm when it is defined in terms of factor graphs, and as Loopy Belief Propagation when defined in terms of Bayesian Networks.
References
Bishop C (2006) Pattern recognition and machine learning. Springer, New York
Boyen X, Koller D (1998) Tractable inference for complex stochastic processes. In: Proceedings of uncertainty in AI. Morgan Kaufmann, San Francisco, pp 33–42
Bui T, Poel M, Nijholt A, Zwiers J (2009) A tractable hybrid DDN-POMDP approach to affective dialogue modeling for probabilistic frame-based dialogue systems. Nat Lang Eng 15(2):273–307
Henderson J, Lemon O (2008) Mixture model POMDPs for efficient handling of uncertainty in dialogue management. In: Proceedings of ACL/HLT. Association for Computational Linguistics, Ohio, pp 73–76
Jensen F (2001) Bayesian networks and decision graphs. Statistics for engineering and information science. Springer, Heidelberg
Kschischang F, Frey B, Loeliger H (2001) Factor graphs and the sum-product algorithm. IEEE Trans Inf Theory 47:498–519
Murphy K (2002) Dynamic Bayesian networks: representation, inference and learning. PhD thesis, Computer Science Division, University of California, Berkeley
Pearl J (1986) Fusion, propagation, and structuring in belief networks. Artif Intell 29(3):241–288
Williams JD (2007a) Applying POMDPs to dialog systems in the troubleshooting domain. In: Proceedings of the HLT/NAACL workshop on bridging the gap: academic and industrial research in dialog technology
Williams JD (2007b) Using particle filters to track dialogue state. In: Proceedings of ASRU
Williams JD, Poupart P, Young S (2005) Factored partially observable Markov decision processes for dialogue management. In: Proceedings of the IJCAI workshop on knowledge and reasoning in practical dialog systems
Young S, Gasic M, Keizer S, Mairesse F, Schatzmann J, Thomson B, Yu K (2009) The hidden information state model: a practical framework for POMDP-based spoken dialogue management. Comput Speech Lang. ISSN:08852308
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2013 Springer-Verlag London
About this chapter
Cite this chapter
Thomson, B. (2013). Maintaining State. In: Statistical Methods for Spoken Dialogue Management. Springer Theses. Springer, London. https://doi.org/10.1007/978-1-4471-4923-1_3
Download citation
DOI: https://doi.org/10.1007/978-1-4471-4923-1_3
Published:
Publisher Name: Springer, London
Print ISBN: 978-1-4471-4922-4
Online ISBN: 978-1-4471-4923-1
eBook Packages: EngineeringEngineering (R0)