L5 Flashcards
1
Q
When do generalized MDPs converge?
A
Add note about non expansions
2
Q
Why does Q-Learning converge.
A
a
3
Q
What is Convergence
A
a
4
Q
What does Q-learning converges to?
A
Q*
5
Q
Non Expansion
A
a
6
Q
Contraction Mapping
A
a
7
Q
Generalized MDP
A
a
8
Q
Control within TD
A
Action chosen by the learner.
9
Q
List types of non expansions.
A
- Order Statistics (min,max) - Fixed convex combinations
