Deterministic Implementations for Reproducibility in Deep Reinforcement Learning.
Prabhat NagarajanGarrett WarnellPeter StonePublished in: CoRR (2018)
Keyphrases
- reinforcement learning
- deterministic domains
- learning algorithm
- function approximation
- reinforcement learning algorithms
- markov decision processes
- partially observable domains
- black box
- optimal policy
- optimal control
- state space
- multi agent
- machine learning
- partial observability
- control problems
- deep learning
- robotic control
- temporal difference learning
- learning agents
- temporal difference
- model free
- learning problems
- learning process
- fully observable
- relational reinforcement learning
- data mining
- mobile robot
- efficient implementation