Buch, Englisch, 119 Seiten, Format (B × H): 155 mm x 235 mm, Gewicht: 2058 g
An end-to-end approach
Buch, Englisch, 119 Seiten, Format (B × H): 155 mm x 235 mm, Gewicht: 2058 g
Reihe: SpringerBriefs in Speech Technology
ISBN: 978-3-319-26198-0
Verlag: Springer International Publishing
This book discusses the Partially Observable Markov Decision Process (POMDP) framework applied in dialogue systems. It presents POMDP as a formal framework to represent uncertainty explicitly while supporting automated policy solving. The authors propose and implement an end-to-end learning approach for dialogue POMDP model components. Starting from scratch, they present the state, the transition model, the observation model and then finally the reward model from unannotated and noisy dialogues. These altogether form a significant set of contributions that can potentially inspire substantial further work. This concise manuscript is written in a simple language, full of illustrative examples, figures, and tables.
Zielgruppe
Research
Autoren/Hrsg.
Fachgebiete
- Technische Wissenschaften Elektronik | Nachrichtentechnik Nachrichten- und Kommunikationstechnik Signalverarbeitung
- Mathematik | Informatik EDV | Informatik Informatik Mensch-Maschine-Interaktion
- Geisteswissenschaften Sprachwissenschaft Computerlinguistik, Korpuslinguistik
- Mathematik | Informatik EDV | Informatik Informatik Künstliche Intelligenz Spracherkennung, Sprachverarbeitung
Weitere Infos & Material
1 Introduction.- 2 A few words on topic modeling.- 3 Sequential decision making in spoken dialog management.- 4 Learning the dialog POMDP model components.- 5 Learning the reward function.- 6 Application on healthcare dialog management.- 7 Conclusions and future work.