A simulation based algorithm for optimal quantization in non-linear/non-Gaussian state-space models.
Vladislav B. TadicAmaud DoucetPublished in: ACC (2004)
Keyphrases
- dynamic programming
- optimal solution
- state space
- state variables
- preprocessing
- exhaustive search
- learning algorithm
- probabilistic model
- worst case
- computational complexity
- search space
- k means
- cost function
- np hard
- closed form
- heuristic search
- simulated annealing
- parameter estimation
- optimization algorithm
- globally optimal
- convergence rate
- monte carlo
- detection algorithm
- computational cost
- objective function
- belief state
- learned models
- evolutionary algorithm
- search algorithm
- reinforcement learning
- clustering algorithm