k-Nearest Neighbor Monte-Carlo Control Algorithm for POMDP-Based Dialogue Systems.
Fabrice LefèvreMilica GasicFilip JurcícekSimon KeizerFrançois MairesseBlaise ThomsonKai YuSteve J. YoungPublished in: SIGDIAL Conference (2009)
Keyphrases
- control algorithm
- monte carlo
- dialogue system
- k nearest neighbor
- policy evaluation
- knn
- markov chain
- nearest neighbor
- finite state
- natural language
- control strategy
- control system
- reinforcement learning
- control method
- dynamical systems
- state space
- text categorization
- partially observable markov decision process
- partially observable markov decision processes
- temporal difference
- partially observable
- variance reduction
- support vector machine
- feature selection
- belief state
- particle filter
- support vector machine svm
- text classification
- monte carlo tree search
- markov decision process
- optimal policy
- human users
- markov decision processes
- multi agent
- neural network
- infinite horizon
- real time
- least squares
- machine learning