Learn Continuously, Act Discretely: Hybrid Action-Space Reinforcement Learning For Optimal Execution.
Feiyang PanTongzhe ZhangLing LuoJia HeShuoling LiuPublished in: CoRR (2022)
Keyphrases
- action space
- reinforcement learning
- state space
- function approximators
- markov decision processes
- continuous state spaces
- state and action spaces
- control policies
- dynamic programming
- continuous state
- real valued
- stochastic processes
- optimal control
- function approximation
- state information
- reinforcement learning methods
- state action
- optimal solution
- action selection
- markov decision process
- learning algorithm
- single agent
- reinforcement learning algorithms
- optimal policy
- learning agent
- domain independent
- markov chain
- markov random field
- control policy
- partially observable
- infinite horizon
- graphical models
- markov decision problems