I’d like to ask, since according to documentation they are exactly the same: is there any difference between those 2 environments? I would expect the dense one to give denser rewards - so every time the agent acquires a rewarding item -, and the other one to only reward on the first occurence of acquiring an item. However that is not the case. So what is the difference, if there is any?
The difference between ObtainDiamondVectorObf and ObtainDiamondDenseVectorObf
Dense: During an episode the agent is rewarded every time it obtains an item in the requisite item hierarchy to obtaining a diamond.
Not dense: Only once per item.
Full description on environments can be found at:
That’s what I thought, however even in the docs you linked the following paragraph is present in both of the environment’s descriptions:
During an episode the agent is rewarded every time it obtains an item in the requisite item hierarchy to obtaining a diamond. The rewards for each item are given here:
Ah, I see. It should be typo mostly in MineRL documentation.
That’s an error with the documentation - I thought we had fixed that but we must have missed a section, sorry!