search

AI4Finance-Foundation / ElegantRL

Massively Parallel Deep Reinforcement Learning. 🔥
https://ai4finance.org
Other
3.67k stars 843 forks source link

Recurrent Neural Network and lookback window #42

Open yosoul93 opened 3 years ago

yosoul93 commented 3 years ago

How can this project implement a Recurrent Neural Network? Because all stock or crypto are time-series and lookback is a very important aspect of trading

Yonv1943 commented 3 years ago

It is difficult to add RNN (LSTM or GRU) for Reinforcement Learning.

The current more mature solution is the unrolled RNN for RL solution of R2D2. We plan to add R2D2 to ElegantRL.

We have fully upgraded ElegantRL and now supports multiple GPU training (1~8 GPU). I'm sorry that we have been busy developing the 80 GPU version (Cloud platform) of ElegantRL, and we were unable to reply to you in time.

Yonv1943 commented 3 years ago

Forgot to say, it is easier to add the lookback window to ElegantRL. We have designed a ReplayBuffer suitable for the lookback window. This function will be added when the time is right.