Difference Based Metrics for Deep Reinforcement Learning Algorithms.
Bernardo Augusto Godinho de OliveiraCarlos Augusto Paiva da Silva MartinsFlávia Magalhães Freitas FerreiraLuís Fabrício Wanderley GóesPublished in: IEEE Access (2019)
Keyphrases
- reinforcement learning algorithms
- reinforcement learning
- markov decision processes
- state space
- model free
- reinforcement learning problems
- temporal difference
- learning algorithm
- reinforcement learning methods
- eligibility traces
- function approximation
- reward function
- stochastic games
- dynamic environments
- policy search
- partially observable environments
- reward shaping
- optimal policy
- particle swarm optimization
- artificial neural networks
- data mining
- multiagent reinforcement learning