114 |
114 |
%\end{enumerate}
|
115 |
115 |
\end{proposition}
|
116 |
116 |
|
117 |
|
\nb{TODO: extend with $\mu$s.}
|
|
117 |
%\nb{TODO: extend with $\mu$s.}
|
118 |
118 |
|
119 |
119 |
|
120 |
|
\nb{Remember polymax bounded checking lemma! Quite important. Also need to bear this in mind when adding functions.}
|
|
120 |
\nb{Remember polymax bounded checking lemma! Quite important. Also need to bear this in mind when adding functions.}
|
|
121 |
|
|
122 |
Now, in order to characterize $\fph$ and its subclasses $\fphi{i}$ we consider Bellantoni's function algebra $\mubc$, defined by using predicative minimization:
|
|
123 |
\begin{definition}[$\mubc$ and $\mubc^i$ programs]
|
|
124 |
We define:
|
|
125 |
\begin{itemize}
|
|
126 |
\item The class $\mubc$ is the set of functions defined as the BC functions, but with the following additional generation rule:
|
|
127 |
|
|
128 |
8. Predicative minimization. If $h$ is in $\mubc$, then so is $f$ defined by
|
|
129 |
|
|
130 |
$f(\vec u; \vec x):= \begin{cases}
|
|
131 |
s_1(\mu y.h(\vec u; \vec x, y)\mod 2 = 0)& \mbox{ if there exists such a $y$,} \\
|
|
132 |
& \mbox{ otherwise,}
|
|
133 |
\end{cases}
|
|
134 |
$
|
|
135 |
|
|
136 |
where $\mu y.h(\vec u; \vec x , y)\mod 2 = 0$ is the least $y$ such that the equality holds.
|
|
137 |
|
|
138 |
If $\Phi$ is a class of functions, we denote by $\mubc(\Phi)$ the class obtained as $\mubc$ but adding $\Phi$ to the set of initial functions.
|
|
139 |
|
|
140 |
\item For each $i\geq 0$, $\mubc^i$ is the set of $\mubc$ functions obtained by at most $i$ applications of predicative minimization. So $\mubc^0=BC$ and $\mubc =\cup_i \mubc^i$.
|
|
141 |
\end{itemize}
|
|
142 |
\end{definition}
|
|
143 |
One then obtains:
|
|
144 |
\begin{theorem}[\cite{BellantoniThesis}]
|
|
145 |
The class of functions representable by $\mubc$ programs is $\fph$, and for each $i$ the class representable by $\mubc^i$ programs is $\fphi{i}$.
|
|
146 |
\end{theorem}
|
|
147 |
We will also need an intermediary lemma, also proved in \cite{BellantoniThesis}.
|
|
148 |
\begin{definition}
|
|
149 |
A function $f(\vec u; \vec x)$ is \textit{polymax bounded} if there exists a polynomial $q$ such that,
|
|
150 |
for any $\vec u$ and $\vec x$ one has:
|
|
151 |
|
|
152 |
%$\size{u}$
|
|
153 |
$$ \size{f(\vec u; \vec x)} \leq q(\size{u_1} , \dots , \size{u_k}) + \max_j \size{x_j}.$$
|
|
154 |
\end{definition}
|
|
155 |
% \begin{lemma}
|
|
156 |
% Let $\Phi$ be a class of polymax bounded functions. If $f$ is in $\mubc(\Phi)$
|
|
157 |
% \end{lemma}
|
|
158 |
\nb{polychecking lemma still to be added}
|
|
159 |
|