L6 - AAA Flashcards
1
Q
1 / 1-gamma
A
Effective Horizon.
The time to find an optimal policy.
2
Q
What is the relationship of gamma in value iteration?
A
Gamma tells you what your horizon is.
Smaller gamma stays more in the present.
Larger gamma pushes out into the future.
3
Q
What happens with a gamma close to 0?
A
Agent becomes short sighted.
4
Q
In what time does Value Iteration solve MDPs.
A
> than polynomial but it is possible to get close to the optimal policy in polynomial time.
