tex-projects/LaTeX-examples
2
0
Fork 0
mirror of https://github.com/MartinThoma/LaTeX-examples.git synced 2025-04-26 06:48:04 +02:00
Code Activity

Add q-learning and improve value iteration pseudocode

Browse source
This commit is contained in:
Martin Thoma 2016-07-14 15:20:45 +02:00
parent 807b9268d0
commit 001350bae4
6 changed files with 88 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

BIN
source-code/Pseudocode/Value-Iteration/Value-Iteration.png
View file

Binary file not shown.
Side by side Swipe Overlay

Before

Width:  |  Height:  |  Size: 23 KiB

After

Width:  |  Height:  |  Size: 27 KiB

Before After
Before After

7
source-code/Pseudocode/Value-Iteration/Value-Iteration.tex
View file

@ -23,8 +23,8 @@
\Statex Actions $\mathcal{A} = \{1, \dots, n_a\},\qquad A: \mathcal{X} \Rightarrow \mathcal{A}$
\Statex Cost function $g: \mathcal{X} \times \mathcal{A} \rightarrow \mathbb{R}$
\Statex Transition probabilities $f$
% TODO: \alpha?
\Procedure{ValueIteration}{$\mathcal{X}$, $A$, $g$, $f$}
\Statex Learning rate $\alpha \in [0, 1]$, typically $\alpha = 0.1$
\Procedure{ValueIteration}{$\mathcal{X}$, $A$, $g$, $f$, $\alpha$}
\State Initialize $J, J': \mathcal{X} \rightarrow \mathbb{R}_0^+$ arbitrarily
\While{$J$ is not converged}
\State $J' \gets J$
@ -37,9 +37,10 @@
\State $J(x) \gets \min_a \{Q(x, a)\}$
\EndFor
\EndWhile
\Return $J$
\EndProcedure
\end{algorithmic}
\caption{Value Iteration}
\caption{Value Iteration: Learn function $J: \mathcal{X} \rightarrow \mathbb{R}$}
\label{alg:value-iteration}
\end{algorithm}
\end{preview}