Kautenja/playing-mario-with-deep-reinforcement-learning
An implementation of (Double/Dueling) Deep-Q Learning to play Super Mario Bros.
Jupyter NotebookMIT
Issues
- 2
Out of memory error during training
#36 opened by aadeshnpn - 1
Broken Prioritized Experience Replay
#33 opened by Kautenja - 0
ImportError: cannot import name 'wrap'
#40 opened by Luziberto - 1
cannot run "python . -m play -o results/SuperMarioBros-1-4-v0/DeepQAgent/2018-08-18_18-33"
#39 opened by Splaver080 - 1
- 2
Do you have results?
#37 opened by zamlz - 5
Training error
#35 opened by aadeshnpn - 2
Training Killed by OS
#29 opened by Kautenja - 1
- 1
Inefficient render_mode usage in Agent
#32 opened by Kautenja - 1
nocite in paper
#1 opened by Kautenja - 1
Environment
#23 opened by Kautenja - 2
DownsampleEnv metadata
#21 opened by Kautenja - 4
Play Time Limit
#11 opened by Kautenja - 1
Games to Test
#16 opened by Kautenja - 2
Negative Reward for terminal flag
#18 opened by Kautenja - 1
Cite Ohio HPC Center
#20 opened by Kautenja - 2
Reward Schemes
#17 opened by Kautenja - 4
DQN: Optimizer
#10 opened by Kautenja - 1
render_mode in Agent
#14 opened by Kautenja - 1
- 1
DQN: epoch
#8 opened by Kautenja - 1
Savable Replay Queue
#7 opened by Kautenja - 1
Optimize Replay Queue
#6 opened by Kautenja - 1
DQN: discount growth?
#4 opened by Kautenja - 1
Downsampler Class
#5 opened by Kautenja - 1
DQN: parameterize optimizer
#3 opened by Kautenja - 1
Cleanup requirements.txt
#2 opened by Kautenja