Composing Ensembles of Policies with Deep Reinforcement Learning.
Ahmed Hussain QureshiJacob J. JohnsonYuzhe QinByron BootsMichael C. YipPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- markov decision process
- control policies
- state space
- fitted q iteration
- reward function
- function approximation
- reinforcement learning agents
- hierarchical reinforcement learning
- markov decision processes
- partially observable markov decision processes
- reinforcement learning algorithms
- multi agent
- decision trees
- total reward
- policy gradient methods
- temporal difference
- model free
- learning algorithm
- ensemble learning
- markov decision problems
- finite state
- decision problems
- control policy
- continuous state
- macro actions
- ensemble methods
- temporal difference learning
- base classifiers
- learning agent
- infinite horizon
- multi agent reinforcement learning
- state abstraction
- decentralized control
- evaluation function
- multiagent reinforcement learning
- transfer learning
- random forests
- multi class
- neural network
- machine learning