MDP QOL Changes

[Sungonogi]

MDP Quality of Life Changes

Explanation Mechanism
Currently the expected answers contain more than the actual solution. It would be nice if there was a way to expect only the solution (f.e. -2), but still have an explanation somewhere.

• Add a mechanism for explanations

• use it for these two cases

 

Discount Factor formula

• consider writing the given discount factor in the formula

 
Goal State Belief State Calculation
0 is the goal state and apparently the transition model always stays there no matter which action we do.

• explain that somewhere and ideally don't use goal states in this exercise

 
Leaving the grid

• explain somewhere that moving out of the area never works

 
Action Text Wrong

• the action should be "right" here:
  (Looking at the code fixing this in a clean way requires a larger change, because we need to differentiate between the cylic track and the grid exercise)

 
feel free to add more