Title :
Using POMDPS for dialog management
Author_Institution :
Eng. Dept., Cambridge Univ., Cambridge
Abstract :
This paper explains how partially observable Markov decision processes (POMDPs) can provide a principled mathematical framework for modelling the inherent uncertainty in spoken dialog systems. It briefly summarises the basic mathematics and explains why exact optimisation is intractable. It then describes a form of approximation called the Hidden Information State model which does scale and which can be used to build practical systems.
Keywords :
Markov processes; interactive systems; natural language processing; speech processing; POMDPS; dialog management; hidden information state model; partially observable Markov decision processes; principled mathematical framework; spoken dialog systems; Decision making; Decoding; Delay; Engineering management; History; Mathematical model; Mathematics; Speech; State estimation; Uncertainty;
Conference_Titel :
Spoken Language Technology Workshop, 2006. IEEE
Conference_Location :
Palm Beach
Print_ISBN :
1-4244-0872-5
DOI :
10.1109/SLT.2006.326785
