Commit 75457690bb885cc229cb0d4f29f10dd08782a4ce

Authored by Francisco Coelho
1 parent 64dd7c8f
Exists in master

Dealing with sub- and supersets.

code/asp/disj.lp
1 % prob(a) = 0.3 1 % prob(a) = 0.3
2 a ; -a. 2 a ; -a.
3 -b ; c :- a.  
4 \ No newline at end of file 3 \ No newline at end of file
  4 +b ; c :- a.
  5 +
  6 +d ; -d.
  7 +b :- c, d.
5 \ No newline at end of file 8 \ No newline at end of file
text/pre-paper.pdf
No preview for this file type
text/pre-paper.tex
1 \documentclass[a4paper, 12pt]{article} 1 \documentclass[a4paper, 12pt]{article}
2 2
3 \usepackage[x11colors]{xcolor} 3 \usepackage[x11colors]{xcolor}
  4 +%
  5 +\usepackage{tikz}
  6 +\usetikzlibrary{calc}
  7 +%
4 \usepackage{hyperref} 8 \usepackage{hyperref}
5 \hypersetup{ 9 \hypersetup{
6 colorlinks=true, 10 colorlinks=true,
7 linkcolor=blue, 11 linkcolor=blue,
8 } 12 }
9 - 13 +%
10 \usepackage{commath} 14 \usepackage{commath}
11 \usepackage{amsthm} 15 \usepackage{amsthm}
12 \newtheorem{assumption}{Assumption} 16 \newtheorem{assumption}{Assumption}
@@ -155,17 +159,39 @@ Equation \ref{eq:world.fold.superset} results from conditional independence of t @@ -155,17 +159,39 @@ Equation \ref{eq:world.fold.superset} results from conditional independence of t
155 Stable models are conditionally independent, given their total choices. 159 Stable models are conditionally independent, given their total choices.
156 \end{assumption} 160 \end{assumption}
157 161
158 -Consider the stable models $ab, ac$ from the example above. They result from the clause $b \vee c \leftarrow a$ and the total choice $a$. These formulas alone impose no relation between $b$ and $c$ (given $a$), so none should be assumed. Dependence relations are discussed in Subsection (\ref{subsec:dependence}). 162 +Consider the stable models $ab, ac$ from the example above. They result from the clause $b \vee c \leftarrow a$ and the total choice $a$. These formulas alone impose no relation between $b$ and $c$ (given $a$), so none should be assumed. Dependence relations are further discussed in Subsection (\ref{subsec:dependence}).
159 163
160 % SUBSET 164 % SUBSET
  165 +\hrule
161 166
  167 +\bigskip
162 I'm not sure about what to say here.\marginpar{todo} 168 I'm not sure about what to say here.\marginpar{todo}
163 169
  170 +My first guess was
164 \begin{equation*} 171 \begin{equation*}
165 \pr{W = w \given C = c} = \sum_{s \supset w}\pr{S = s \given C = c}. 172 \pr{W = w \given C = c} = \sum_{s \supset w}\pr{S = s \given C = c}.
166 \end{equation*} 173 \end{equation*}
167 174
168 -But! $\pr{W = w \given C = c}$ already separates $\pr{W}$ into \textbf{disjoint} 175 +$\pr{W = w \given C = c}$ already separates $\pr{W}$ into \textbf{disjoint} events!
  176 +
  177 +Also, I am assuming that stable models are independent.
  178 +
  179 +This would entail $p(w) = p(s_1) + p(s_2) - p(s_1)p(s_2)$ \textit{if I'm bound to set inclusion}. But I'm not. I'm defining a relation
  180 +
  181 +Also, if I set $p(w) = p(s_1) + p(s_2)$ and respect the laws of probability, this entails $p(s_1)p(s_2) = 0$.
  182 +
  183 +So, maybe what I want is (1) to define the cover $\hat{w} = \cup_{s \supset w} s$
  184 +
  185 +\begin{equation*}
  186 + \pr{W = w \given C = c} = \sum_{s \supset w}\pr{S = s \given C = c} - \pr{W = \hat{w} \given C = c}.
  187 +\end{equation*}
  188 +
  189 +But this doesn't works, because we'd get $\pr{W = a \given C = a} < 1$.
  190 +%
  191 +
  192 +%
  193 +\bigskip
  194 +\hrule
169 195
170 % INDEPENDENCE 196 % INDEPENDENCE
171 197
@@ -188,9 +214,9 @@ $ @@ -188,9 +214,9 @@ $
188 $. 214 $.
189 215
190 216
191 -The interesting case is the subtree of the total choice $ac$. Notice that no stable model $s$ contains $adc$ because (1) $adb$ is a stable model and (2) no stable model contains $adc$, because if $adc \subset s$ then $b \in s$ and $adb \subset s$. 217 +The interesting case is the subtree of the total choice $ad$. Notice that no stable model $s$ contains $adc$ because (1) $adb$ is a stable model and (2) if $adc \subset s$ then $b \in s$ so $adb \subset s$.
192 218
193 -Following the case of equations (\ref{eq:world.fold.stablemodel}) and (\ref{eq:world.fold.independent}) this sets 219 +Following equations (\ref{eq:world.fold.stablemodel}) and (\ref{eq:world.fold.independent}) this sets
194 \begin{equation*} 220 \begin{equation*}
195 \begin{cases} 221 \begin{cases}
196 \pr{W = adc \given C = ad} = 0,\cr 222 \pr{W = adc \given C = ad} = 0,\cr
@@ -200,7 +226,7 @@ Following the case of equations (\ref{eq:world.fold.stablemodel}) and (\ref{eq:w @@ -200,7 +226,7 @@ Following the case of equations (\ref{eq:world.fold.stablemodel}) and (\ref{eq:w
200 which concentrates all probability mass from the total choice $ad$ in the $adb$ branch, including the node $W = adbc$. This leads to the following cases: 226 which concentrates all probability mass from the total choice $ad$ in the $adb$ branch, including the node $W = adbc$. This leads to the following cases:
201 $$ 227 $$
202 \begin{array}{l|r} 228 \begin{array}{l|r}
203 - x & \pr{x \given C = ad}\ 229 + x & \pr{W = x \given C = ad}\
204 \hline 230 \hline
205 ad & 1 \\ 231 ad & 1 \\
206 adb & 1\\ 232 adb & 1\\
@@ -208,8 +234,22 @@ $$ @@ -208,8 +234,22 @@ $$
208 adbc & 1 234 adbc & 1
209 \end{array} 235 \end{array}
210 $$ 236 $$
211 -so, for $C = ad$, $\pr{b} = 2/3, \pr{c} = 1/3, \pr{b,c} = 1/3 \not= \pr{b}\pr{c}$. 237 +so, for $C = ad$,
  238 +$$
  239 +\begin{aligned}
  240 + \pr{W = b} &= \frac{2}{4} \cr
  241 + \pr{W = c} &= \frac{1}{4} \cr
  242 + \pr{W = bc} &= \frac{1}{4} \cr
  243 + &\not= \pr{W = b}\pr{W = c}
  244 +\end{aligned}
  245 +$$
  246 +\textit{i.e.} the events $W = b$ and $W = c$ are dependent and that dependence results directly from the segment $0.2::d, b \leftarrow c \wedge d$ in the specification.
  247 +
212 248
  249 +%
  250 +
  251 +%
  252 +\hrule
213 \begin{quotation}\note{Todo} 253 \begin{quotation}\note{Todo}
214 254
215 Prove the four world cases (done), support the product (done) and sum (tbd) options, with the independence assumptions. 255 Prove the four world cases (done), support the product (done) and sum (tbd) options, with the independence assumptions.