Lunar-Lander with DQN DDQN and Duel DQN

[proud ridge]

Hello everyone, I implemented Lunar Lander with DQN and DDQN. Now I also want to add Duel DQN and compare between the 3. Feel free to contribute ! ❤️ link to repo : https://github.com/femaleprog/lunar-lunder--D-DQN/
#🦾・reinforcement-learning

[brisk halo]

Like the notebooks! As a starter for looking at methods of exploration other than epsilon-greedy, have a look at noisy networks: https://arxiv.org/abs/1706.10295. I think it's the best DQN trick!