An Online Algorithm for Applying Reinforcement Learning to Handle Ambiguity in Spoken Dialogues.
Fangju WangKyle SweglesPublished in: TAMC (2009)
Keyphrases
- reinforcement learning
- learning algorithm
- dynamic programming
- computational cost
- experimental evaluation
- objective function
- cost function
- stochastic approximation
- times faster
- matching algorithm
- detection algorithm
- linear programming
- np hard
- probabilistic model
- high accuracy
- computationally efficient
- significant improvement
- k means
- search space
- improved algorithm
- preprocessing
- least squares
- worst case
- particle swarm optimization
- tree structure
- recognition algorithm