The Black Box Effect Is Completely Absent in Reinforcement Learning

Pipeline inverse reinforcement learning is directly enforcement learning but both elements in our constructions do not involve neural nets okay so they involve just linear algebra. All which is needed to implement it is not the deep network packages but rather standard you know linear algebra packages like numpyOkay and that's why i call it white box because it's literally what it does is determined by formulas which are written right  and there are no hidden parameters rightSo all parameters that mean for our right there in right and you can check every step in these formulas with doty land under the hoodRight now our design has a black box algorithm withdoty land is standard school under the hood right nowOur general structure