Training Reinforcement Neurocontrollers Using the Polytope Algorithm
Aristidis LikasIsaac E. LagarisPublished in: CoRR (1998)
Keyphrases
- matching algorithm
- optimization algorithm
- dynamic programming
- experimental evaluation
- detection algorithm
- computational cost
- learning algorithm
- preprocessing
- high accuracy
- convex hull
- cost function
- simulated annealing
- segmentation algorithm
- computational complexity
- significant improvement
- np hard
- knapsack problem
- linear programming
- expectation maximization
- times faster
- optimal solution
- extreme points
- classification algorithm
- theoretical analysis
- search space
- lower bound
- training data