Open Dronie opened 1 month ago
hey thanks for opening this :) we're busy with neurips rebuttals atm but will come back to this after!
Hey! just took a look and it looks good but could you add a flag for shared vs individual rewards? @Dronie
Hey! just took a look and it looks good but could you add a flag for shared vs individual rewards? @Dronie
Done :)
Slightly changed the way observations, rewards, dones and infos are processed/returned to make coin game compatible with iql_rnn (and potentially others but this is the only algorithm I tested). Also added line 352 which allows for shared rewards (each agent's reward becomes the sum of their individual rewards) but not yet integrated in a nice way.
See plots for training results (using same configs as in ql_rnn_mpe):![image](https://github.com/user-attachments/assets/e14a2075-cff6-4757-aab2-3992380aeb3e)
Also see gifs of trained policies in action (10 episodes each): Default (individual) rewards:![individual_rewards](https://github.com/user-attachments/assets/627e1c97-2df5-43e0-89e1-94cfb71700c9)
Shared rewards:![shared_rewards](https://github.com/user-attachments/assets/3e94977b-188a-4f3e-8ec3-a892f7be28b5)
NB: have not done unit tests as this is only a minor change to an existing environment - let me know if this is an issue!