Model-based Safe Deep Reinforcement Learning via a Constrained Proximal Policy Optimization Algorithm.
Ashish Kumar JayantShalabh BhatnagarPublished in: CoRR (2022)
Keyphrases
- optimization algorithm
- reinforcement learning
- optimal policy
- model free
- policy search
- multi objective
- action selection
- optimization method
- markov decision process
- reinforcement learning problems
- partially observable environments
- reinforcement learning algorithms
- state and action spaces
- differential evolution
- partially observable
- evolutionary multi objective
- policy iteration
- function approximators
- markov decision processes
- optimization strategy
- markov decision problems
- reward function
- partially observable markov decision processes
- function approximation
- particle swarm optimization pso
- control policy
- action space
- actor critic
- policy gradient
- learning algorithm
- average reward
- state action
- state space
- global optima
- dynamic programming
- agent learns
- hybrid optimization algorithm
- agent receives
- multiobjective optimization
- genetic programming
- search space
- objective function
- neural network