This repository has been archived by the owner on Mar 25, 2022. It is now read-only.
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Jeffrey Kegler
committed
Feb 25, 2016
1 parent
f1c710d
commit 150bd6d
Showing
2 changed files
with
158 additions
and
79 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,100 +1,89 @@ | ||
\begin{theorem} | ||
\ttitle{Leo indirect effect properties} | ||
\label{t:leo-indirect-effect-props} | ||
We make the following assumptions: | ||
\begin{subequations} | ||
\renewcommand{\theequation}{A\arabic{equation}} | ||
\ttitle{Medial EIM as a cause} | ||
\label{t:medial-eim-as-cause} | ||
Assume that | ||
\begin{subequation} | ||
\rewew | ||
\begin{align} | ||
\label{eq:leo-indirect-effect-props-asm-1} | ||
\label{t:medial-eim-as-cause-asm-1} | ||
& \myparbox{% | ||
\Veim{cuz} is a valid EIM. | ||
There exist two valid EIM's, | ||
such that \Veim{cuz} | ||
is the cause of \Veim{eff}. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-asm-2} | ||
\label{t:medial-eim-as-cause-asm-1} | ||
& \myparbox{% | ||
\Vleo{l} is an instantiated Leo memo. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-asm-3} | ||
& \myparbox{% | ||
\Vleo{l} does not match \Veim{cuz}. | ||
\Veim{cuz} is incomplete. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-asm-4} | ||
\label{t:medial-eim-as-cause-asm-1} | ||
& \myparbox{% | ||
\Vleo{l} quasi-matches \Veim{cuz}. | ||
\Veim{cuz} is not a prediction. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-asm-5} | ||
& \myparbox{% | ||
\Vleo{l} matches \Veim{cuz}. | ||
} | ||
\end{align} | ||
\end{subequations} | ||
|
||
If these assumptions hold, then | ||
\begin{subequations} | ||
\renewcommand{\theequation}{R\arabic{equation}} | ||
\end{subequation} | ||
Given these assumptions, then | ||
\begin{subequation} | ||
\rewew | ||
\begin{align} | ||
\label{eq:leo-indirect-effect-props-req-0-5} | ||
& \myparbox{% | ||
\Vleo{l} has exactly one valid basis, \Veim{bas}. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-req-1} | ||
& \myparbox{% | ||
\Veim{cuz} and \Vleo{l} have exactly one valid direct Leo effect, | ||
call it \Veim{eff}. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-req-2} | ||
\label{t:medial-eim-as-cause-req-1} | ||
& \myparbox{% | ||
\Veim{cuz} is a top-down cause of \Veim{eff}. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-req-3} | ||
& \myparbox{% | ||
\Veim{eff} is the only effect of \Veim{cuz}. | ||
} \\ | ||
& \myparbox{% | ||
The only bottom-up cause of | ||
\Veim{eff} is ethereal. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-req-4} | ||
& \myparbox{% | ||
\Veim{cuz} is the only top-down cause | ||
of \Veim{eff}. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-req-5} | ||
& \myparbox{% | ||
$\Rule{\Veim{eff}} = \Rule{\Veim{cuz}}$. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-req-6} | ||
& \myparbox{% | ||
$\Right{\Veim{eff}} = \Right{\Veim{cuz}}$. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-req-7} | ||
& \myparbox{% | ||
\Veim{cuz} is quasi-complete. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-req-8} | ||
& \myparbox{% | ||
\Veim{cuz} is incomplete. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-req-9} | ||
& \myparbox{% | ||
\Veim{eff} is quasi-complete. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-req-10} | ||
& \myparbox{% | ||
\Veim{cuz} is a silo cause of \Veim{eff}. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-req-11} | ||
\label{t:medial-eim-as-cause-req-2} | ||
& \myparbox{% | ||
\Veim{eff} is the only silo effect of \Veim{cuz}. | ||
} \\ | ||
\label{eq:leo-indirect-effect-props-req-12} | ||
& \myparbox{% | ||
\Vleo{l} matches \Veim{eff}. | ||
\Veim{cuz} is not the bottom-up cause of any effect. | ||
} | ||
\intertext{% | ||
and one of the following two | ||
cases hold: | ||
} | ||
\label{t:medial-eim-as-cause-req-3a} | ||
& \myparbox{% | ||
\Veim{eff} is a confirmed EIM. | ||
and the effect of the cause-pair | ||
$[ \Veim{cuz}, \Vinst{up} ]$, | ||
where \Vinst{up} is a parse instance. | ||
} \\ | ||
\label{t:medial-eim-as-cause-req-3b} | ||
& \myparbox{ | ||
\Veim{eff} is a predicted EIM | ||
whose top-down cause is \Veim{cuz} | ||
and whose bottom-up cause is ethereal. | ||
} | ||
\end{align} | ||
\end{subequations} | ||
\end{subequation} | ||
\end{theorem} | ||
|
||
\begin{proof} | ||
TODO | ||
We proceed by cases. | ||
|
||
\textbf{Top-down cause of a prediction}: | ||
If \Veim{eff} is a prediction, | ||
then its bottom-up cause is ethereal | ||
\becuz{} TODO. | ||
Therefore, \Veim{cuz} must be its top-down cause. | ||
|
||
\textbf{Bottom-up cause of a prediction}: | ||
If \Veim{eff} has no predot symbol, | ||
then its bottom-up cause is ethereal | ||
\dref[causes of a predicted EIM]{def:causes-predicted}. | ||
\Veim{cuz} is clearly not ethereal, | ||
so that \Veim{cuz} is not the bottom-up cause of \Veim{eff}. | ||
|
||
\textbf{Top-down cause of a confirmation}: | ||
If \Veim{eff} has a predot symbol, | ||
its causes are in a pair of matching causes | ||
\dref[causes of a confirmed EIM]{def:causes-confirmed}. | ||
TODO: finished? | ||
|
||
\textbf{Bottom-up cause of a confirmation}: | ||
The bottom-up cause of a confirmation must have a | ||
symbolic equivalent | ||
\becuz{} TODO. | ||
\Veim{cuz} is incomplete and therefore has no | ||
symbolic equivalent | ||
\becuz{} TODO. | ||
|
||
TODO: finished? | ||
\end{proof} | ||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters