Description

Comparison of DQN and its variants.

Usage

1. Specify hyper-parameters in config.py.
2. Run ./run.sh $device $algorithm for training with three random seeds.
3. Visualize the results by python plot.py *log*.

Result on Breakdown

Cite

@book{deepRL-2020,
 title={Deep Reinforcement Learning: Fundamentals, Research, and Applications},
 editor={Hao Dong, Zihan Ding, Shanghang Zhang},
 author={Hao Dong, Zihan Ding, Shanghang Zhang, Hang Yuan, Hongming Zhang, Jingqing Zhang, Yanhua Huang, Tianyang Yu, Huaqing Zhang, Ruitong Huang},
 publisher={Springer Nature},
 note={\url{http://www.deepreinforcementlearningbook.org}},
 year={2020}
}