Replay Memory without History

[etotheipluspi]

Should we automatically check if the user is requesting only the current observation as input to the DQN? We can then decide to use a more efficient memory sampling scheme than the one that samples histories.

The other option would be for the user to pick the memory that they want (with history sampling functionality or without).

[yegortk]

I think it's best if we have one general sampling scheme without special cases - no history can be the default, i.e. (1,0,0).