A new algorithm to track dynamic goal position in Q-learning.
Soumishila MitraDhrubojyoti BanerjeeAmit KonarRamadoss JanarthananPublished in: HIS (2012)
Keyphrases
- learning algorithm
- times faster
- dynamic programming
- optimization algorithm
- recognition algorithm
- bucket brigade
- theoretical analysis
- objective function
- computational complexity
- detection algorithm
- experimental evaluation
- segmentation algorithm
- monte carlo
- worst case
- np hard
- convergence rate
- k means
- search space
- preprocessing
- stochastic approximation
- improved algorithm
- computational cost
- expectation maximization
- particle swarm optimization
- high accuracy
- neural network
- cost function
- significant improvement
- reinforcement learning
- similarity measure
- machine learning