Proximal Policy Optimization-Based Reinforcement Learning and Hybrid Approaches to Explore the Cross Array Task Optimal Solution.
Samuel CoreccoGiorgia AdorniLuca Maria GambardellaPublished in: Mach. Learn. Knowl. Extr. (2023)
Keyphrases
- hybrid approaches
- reinforcement learning
- optimal solution
- optimal policy
- action selection
- policy iteration
- policy search
- optimization problems
- markov decision process
- np hard
- optimization process
- state and action spaces
- global optimization
- linear programming
- evolutionary algorithm
- objective function
- policy evaluation
- policy gradient
- function approximation
- metaheuristic
- optimization algorithm
- multi agent
- markov decision problems
- average reward
- knapsack problem
- approximate dynamic programming
- reinforcement learning problems
- learning algorithm
- working set
- partially observable environments
- machine learning
- state action
- function approximators
- action space
- iterative procedure
- partially observable markov decision processes
- reinforcement learning algorithms
- reward function
- model free
- optimization method