Location
Badges
Activity
Ratings Progression
Challenge Categories
Challenges Entered
Measure sample efficiency and generalization in reinforcement learning using procedurally generated environments
Latest submissions
See Allgraded | 70122 |
Multi-Agent Reinforcement Learning on Trains
Latest submissions
Multi-agent RL in game environment. Train your Derklings, creatures with a neural network brain, to fight for you!
Latest submissions
Multi-Agent Reinforcement Learning on Trains
Latest submissions
Participant | Rating |
---|
Participant | Rating |
---|
Dr. Derks Mutant Battlegrounds
Random items generation question
Over 3 years agoI think empty slots are bad idea.
Removing empty slots will solve half of the issues.
Random items generation question
Over 3 years agoReviewing observations after reset I saw situation where one of my units had only one item.
Also I saw they had same items at least on the tail.
Random items generation question
Over 3 years agoPossible solution is to remove very weak combinations.
Left only 10-20 possible combinations or something like this.
Random items generation question
Over 3 years agoLooks like when we reset DerkEnv we generate a new item set everytime. Which is good.
But Iβve found that my bots sometimes didnβt get all 3 items.
Is it possible during evaluation? If not I believe it should be fixed. And I think it should be configurable in the DerkEnv.reset method. So every reset I can choose what items to use.
NeurIPS 2020: Procgen Competition
Could you confirm that distribution_mode is hard?
Over 4 years agoThank you very much for the answer.
I misunderstood that: β In both cases, agents will be evaluated on the full distribution of levels.β
Could you confirm that distribution_mode is hard?
Over 4 years agoI was pretty sure it is hard but in starter kit sources I saw:
βdistribution_modeβ : βeasyβ # What variant of the levels to use, the options are βeasyβ, βhardβ, βextremeβ, βmemoryβ, βexplorationβ. All games support βeasyβ and βhardβ, while other options are game-specific. The default is βhardβ. Switching to βeasyβ will reduce the number of timesteps required to solve each game and is useful for testing or when working with limited compute resources. NOTE : During the evaluation phase (rollout), this will always be overriden to βeasyβ
Challenge announcement | GPU submissions, build & run logs, and more
Over 3 years agocan confirm. We got the same issues.
Also it feels like that if I do same submission twice I get different results( it could be an issue on my side too).
Also I have suggestion that games should be symmetric. It means if you generated 1000 random combinations vs other 1000 combinations then sides should be swapped and play again the same combinations but for different sides.