copy latest revisions

adzcai · Aug 30, 2024 · d25532d · d25532d
1 parent 7af8b08
commit d25532d
Show file tree

Hide file tree

Showing 5 changed files with 716 additions and 230 deletions.
diff --git a/book/bandits.md b/book/bandits.md
@@ -950,5 +950,3 @@ regret bound. The full details of the analysis can be found in Section 3 of {cit
 +++
 
 ## Summary
-
-
diff --git a/book/control.md b/book/control.md
@@ -971,6 +971,7 @@ Local linearization might only be accurate in a small region around the
 point of linearization.
 :::
 
+(iterative_lqr)=
 ### Iterative LQR
 
 To address these issues with local linearization, we'll use an iterative

diff --git a/book/imitation_learning.md b/book/imitation_learning.md
@@ -137,5 +137,6 @@ def dagger_pseudocode(
     return π
 ```
 
+How well does DAgger perform?
 
-
+<!-- TODO -->
diff --git a/book/mdps.md b/book/mdps.md
@@ -353,6 +353,8 @@ policy for that state and $0$ otherwise. In this case, the only
 randomness in sampling trajectories comes from the initial state
 distribution $\mu$ and the state transitions $P$.
 
++++
+
 ### Value functions
 
 The main goal of RL is to find a policy that maximizes the average total
Original file line number	Diff line number	Diff line change
Expand Up		@@ -950,5 +950,3 @@ regret bound. The full details of the analysis can be found in Section 3 of {cit
		+++

		## Summary