Dynamic ensemble learning based on RL and multi-objective optimization. Deep reinforcement learning and NSGA2 are combined to realize dynamic ensemble learning.
This code is an open source part of the paper 《A dynamic ensemble deep deterministic policy gradient recursive network for spatiotemporal traffic speed forecasting in an urban road network》. For commercial reasons, the code and data in the original paper cannot be open-source. We use matlab to reconstruct the reinforcement learning and multi-objective optimization algorithms in the above paper. And the open source code has been modified to some extent to comply with the relevant regulations.
If this code can help you, please cite the following paper
@article{mi2022dynamic,
title={A dynamic ensemble deep deterministic policy gradient recursive network for spatiotemporal traffic speed forecasting in an urban road network},
author={Mi, Xiwei and Yu, Chengqing and Liu, Xinwei and Yan, Guangxi and Yu, Fuhao and Shang, Pan},
journal={Digital Signal Processing},
volume={129},
pages={103643},
year={2022},
publisher={Elsevier}
}