FinRL-podracer: high performance and scalable deep reinforcement learning for quantitative finance.
Zechu LiXiao-Yang LiuJiahao ZhengZhaoran WangAnwar WalidJian GuoPublished in: ICAIF (2021)
Keyphrases
- reinforcement learning
- data intensive
- function approximation
- data intensive computing
- highly scalable
- reinforcement learning algorithms
- temporal difference
- state space
- computational intelligence
- qualitative and quantitative
- quantitative and qualitative
- markov decision processes
- robotic control
- low latency
- web scale
- machine learning
- financial markets
- learning algorithm
- multi agent
- dynamic programming
- high efficiency
- data sets
- lightweight
- action selection
- markov decision process
- memory efficient
- optimal policy
- quantitative measures
- cost effective
- neural network