Direct heuristic dynamic programming with augmented states.
Jian SunFeng LiuJennie SiShengwei MeiPublished in: IJCNN (2011)
Keyphrases
- dynamic programming
- dynamic programming algorithms
- state space
- linear programming
- lagrangian relaxation
- markov decision problems
- optimal policy
- stereo matching
- neural network
- state variables
- knapsack problem
- greedy algorithm
- markov decision processes
- multiscale
- optimal control
- state transition
- initial state
- dp matching
- coarse to fine
- search algorithm
- reinforcement learning
- beam search
- goal state
- data sets