PHH: Policy-Based Hyper-Heuristic With Reinforcement Learning.
Orachun UdomkasemsubBooncharoen SirinaovakulTiranee AchalakulPublished in: IEEE Access (2023)
Keyphrases
- hyper heuristics
- reinforcement learning
- optimal policy
- policy search
- evolutionary algorithm
- examination timetabling
- markov decision process
- genetic programming
- action selection
- timetabling problem
- function approximation
- function approximators
- constraint satisfaction problems
- control policy
- markov decision processes
- metaheuristic
- partially observable
- action space
- state space
- markov decision problems
- reward function
- difficult problems
- actor critic
- policy iteration
- heuristic search
- reinforcement learning algorithms
- policy gradient
- dynamic programming
- heuristic methods
- model free
- search procedure
- average reward
- graph coloring
- partially observable markov decision processes
- multi objective
- temporal difference
- cutting stock problems
- grasp with path relinking
- neural network
- agent learns
- optimization problems
- special case
- machine learning