A new hybrid method of recurrent reinforcement learning and BiLSTM for algorithmic trading.
Yuling HuangYunlin SongPublished in: J. Intell. Fuzzy Syst. (2023)
Keyphrases
- reinforcement learning
- function approximation
- feed forward
- electronic commerce
- reinforcement learning algorithms
- state space
- recurrent neural networks
- hybrid method
- machine learning
- robotic control
- model free
- optimal policy
- dynamic programming
- simulated annealing
- multi agent
- action selection
- temporal difference
- financial markets
- neural network
- non stationary
- policy search
- foreign exchange
- autonomous learning
- action space
- genetic algorithm
- learning algorithm
- least squares
- markov decision processes
- transfer learning
- monte carlo