Statistiques
| Révision :

root / CSL17 / pv-theories.tex @ 258

Historique | Voir | Annoter | Télécharger (14,92 ko)

1 154 adas
\section{An extension of Bellantoni's theory PV for PH}
2 154 adas
3 157 adas
PVBC+FCA+ safe induction characterises PH?
4 154 adas
5 157 adas
We differ slightly from Bellantoni's theory, using  a mismatch of pvbc and qpvbc.
6 157 adas
This is for simplicity, since
7 157 adas
8 157 adas
\subsection{Base theory}
9 157 adas
10 157 adas
Let us begin with a variation of the base theory of $\pvbci{(0)}$.
11 157 adas
12 156 adas
\begin{definition}
13 157 adas
	[Basic theory]
14 157 adas
	We have two types/sorts: $\normal$ (normal) and $\safe$ (safe).
15 157 adas
	Our language consists of the following:
16 157 adas
	\begin{itemize}
17 157 adas
		\item Normal and safe variables $x^\normal_i , x^\safe_i$. (sometimes write $u$, $x$ , or $\vec u; \vec x$ without superscripts).
18 157 adas
%		\item Constant symbol $0^\safe$.
19 157 adas
%		\item Function symbols $\succ{i}^{\safe \to \safe}$, for $i = 0,1$. (Simply written $\succ i^\safe (;x)$.
20 157 adas
		\item (all initial symbols from BC with their natural typing, plus a recursion scheme. Have the defining axioms too.)
21 157 adas
		\item The following interrelations between safe and normal terms:
22 157 adas
		\begin{itemize}
23 157 adas
			\item For each normal variable $x^\normal$ a safe variable $x^{\normal\safe}$.
24 157 adas
			\item For each normal function symbol $f^\normal$ a safe function symbol $f^{\normal\safe}$.
25 157 adas
			\item For each normal term $t = f^\normal (\vec s ; \vec t)$, we define $t^\safe$ as $f^{\normal\safe} (\vec s ; \vec t)$, and if $t$ is $x^\normal$ then $t^\safe$ as $x^{\normal\safe}$.
26 157 adas
			\item We have the axioms $t^\safe = t$.
27 157 adas
		\end{itemize}
28 157 adas
		\item We have function symbols for the characteristic functions of each quantifier-free predicate. I.e.,
29 157 adas
		\[
30 157 adas
		f_A (\vec u ; \vec x) = 1
31 157 adas
		\quad \ciff \quad
32 157 adas
		A(\vec u ; \vec x)
33 157 adas
		\]
34 157 adas
		where $A$ is quantifier-free. (this is just $\wit{}{}$ for q.f.\ formulae.)
35 156 adas
36 157 adas
		(maybe use $\chi[A]$ for characteristic, to avoid subscripts)
37 157 adas
		\item Open induction, i.e.\ quantifier-free induction, on normal variables:
38 157 adas
		\[
39 157 adas
		\forall \vec u ; \vec x.
40 157 adas
		(A(0) \cand \forall u^\normal . ( A(u) \cimp  A(\succ 0 u ) \cand \forall u^\normal . ( A(u) \cimp  A(\succ 1 u ) ))
41 157 adas
		\cimp \forall u^\normal . A( u ))
42 157 adas
		\]
43 157 adas
		where $\vec u ; \vec x $ are all the free variables of $A(0)$.
44 157 adas
	\end{itemize}
45 156 adas
\end{definition}
46 156 adas
47 157 adas
The following was essentially proved by Bellantoni in his thesis:
48 156 adas
49 157 adas
\begin{theorem}
50 157 adas
	[Bellantoni]
51 157 adas
	$\pvbci{0}$ proves the totality of just the polynomial-time functions.
52 157 adas
\end{theorem}
53 157 adas
54 157 adas
\nb{Notice that induction can be assumed to be simply functional, due to comprehension function symbols. This is more in the spirit of, say, 2nd order subsystems of arithmetic, like in reverse mathematics.}
55 157 adas
56 157 adas
\anupam{Hold on, the above should be fairly immediate, the interesting result is for $\pvbci 1$, which we should perhaps state later.}
57 157 adas
58 157 adas
\subsection{Extensions to theories for the polynomial hierarchy}
59 157 adas
60 157 adas
We now will associate a rank to new function symbols based on the complexity of formulae they characterise.
61 157 adas
62 157 adas
\anupam{Ranks are necessary since we do not have sharply bounded quantifiers to reason about sequences. Check what Krajicek does, probably has the latter.}
63 157 adas
64 157 adas
\begin{definition}
65 157 adas
[Quantifier hierarchy and functional comprehension]
66 157 adas
We mutually define the \emph{safe arithmetical hierarchy} and functions of rank $i$ as follows:
67 157 adas
	\begin{itemize}
68 157 adas
		\item All functions from the basic theory have \emph{rank} 0.
69 157 adas
		\item $\Sigma^\safe_0 = \Pi^\safe_0$ is the set of quantifier-free formulae with just functions symbols of rank $0$.
70 157 adas
		\item The functions symbols of rank $i$ are the characteristic functions of $\Sigma^\safe_i$ formulae.
71 157 adas
		\item $\Sigma^\safe_{i+1}$ is the least class containing:
72 157 adas
		\begin{itemize}
73 157 adas
			\item $\Pi^\safe_i$.
74 157 adas
			\item Atomic formulae $s=t$ where $s$ and $t$ contain only function symbols of rank $\leq i+1$.
75 157 adas
			\item Closed under $\cor , \cand , \exists^\safe$.
76 157 adas
		\end{itemize}
77 157 adas
		\item $\Pi^\safe_{i+1} $ is...
78 157 adas
	\end{itemize}
79 157 adas
\end{definition}
80 157 adas
81 157 adas
\anupam{check the parameters/subscripts above! Aim for:
82 157 adas
83 157 adas
	rank = $\mu$ nesting = quant alt = level - 1
84 157 adas
	}
85 157 adas
86 158 adas
\nb{Wait, check: do we want characteristic functions or functional-comprehension functions? Compare with Krajicek.}
87 157 adas
88 158 adas
89 157 adas
Over the next two sections we will show that the provably total functions of $\pvbci i+1$ are just
90 157 adas
91 156 adas
\subsection{Soundness}
92 156 adas
We show that provably total functions of $\pvbci i$ are in $\fphi i$.
93 156 adas
94 156 adas
The following is our main result:
95 156 adas
\begin{theorem}
96 157 adas
	If $\pvbci {i+1} \proves \forall \vec u^\normal , \vec x^\sigma . \exists \vec y^\sigma. A(\vec u , \vec x , \vec y)$, then there are rank $i $ functions $\vec f (\vec u ; \vec x)$ such that $\pvbci {i+1} \proves\forall \vec u^\normal , \vec x^\sigma . A(\vec u , \vec x , \vec f (\vec u ; \vec x) )  $.
97 156 adas
\end{theorem}
98 156 adas
99 157 adas
\begin{corollary}
100 157 adas
The provably total functions of $\pvbci i+1$ are in $\fphi i$.
101 157 adas
\end{corollary}
102 157 adas
103 156 adas
\begin{definition}
104 156 adas
	[Witness predicate]
105 156 adas
	The \emph{witness predicate} is a $\mubci{}$ program $\wit{\vec a}{A}$, parametrised by variables $\vec a$ and a formula $A$ whose free variables are amongst $\vec a$, defined as follows.
106 156 adas
	If $A$ is a $\Pi_{i}$ formula then:
107 156 adas
	\[
108 156 adas
	\begin{array}{rcl}
109 156 adas
	\wit{\vec u; \vec x}{s=t} (\vec u ; \vec x, w) & \dfn & =(;s,t) \\
110 156 adas
	\smallskip
111 156 adas
	\wit{\vec u; \vec x}{s\neq t} (\vec u ; \vec x, w) & \dfn & \neg (;=(;s,t)) \\
112 156 adas
	\smallskip
113 156 adas
	\wit{\vec u ; \vec x}{A\cor B} (\vec u ; \vec x , w) & \dfn & \cor (; \wit{\vec u , \vec x}{A} (\vec u ; \vec x , w), \wit{\vec u , \vec x}{B} (\vec u ;\vec x, w) ) \\
114 156 adas
	\smallskip
115 156 adas
	\wit{\vec u ; \vec x}{A\cand B} (\vec u ; \vec x , w) & \dfn & \cand(; \wit{\vec u , \vec x}{A} (\vec u ; \vec x , w), \wit{\vec u , \vec x}{B} (\vec u ;\vec x, w) ) \\
116 156 adas
	\smallskip
117 156 adas
	\wit{\vec u ; \vec x}{\exists x^\safe . A(x)} (\vec u ;\vec x,  w) & \dfn & \begin{cases}
118 156 adas
	1 & \exists x^\safe . \wit{\vec u; \vec x, x }{A(x)} (\vec u ;\vec x , x, w) = 1 \\
119 156 adas
	0 & \text{otherwise}
120 156 adas
	\end{cases} \\
121 156 adas
	\smallskip
122 156 adas
	\wit{\vec u; \vec x}{\forall x^\safe . A(x)} (\vec u ;\vec x , w) & \dfn &
123 156 adas
	\begin{cases}
124 156 adas
	0 & \exists x^\sigma. \wit{\vec u ; \vec x , x}{ A(x)} (\vec u; \vec x , x) = 0 \\
125 156 adas
	1 & \text{otherwise}
126 156 adas
	\end{cases}
127 156 adas
	\end{array}
128 156 adas
	\]
129 156 adas
	We now define $\Wit{\vec a}{A}$ for a $\Sigma_{i+1}$-formula $A$ with free variables amongst $\vec a$.
130 156 adas
	\[
131 156 adas
	\begin{array}{rcl}
132 156 adas
	\Wit{\vec u ; \vec x}{A} (\vec u ; \vec x , w) & \dfn & \wit{\vec u ; \vec x}{A} (\vec u ; \vec x)  \text{ if $A$ is $\Pi_i$} \\
133 156 adas
	\smallskip
134 156 adas
	\Wit{\vec u ; \vec x}{A \cor B} (\vec u ; \vec x , \vec w^A , \vec w^B) & \dfn & \cor ( ; \Wit{\vec u ; \vec x}{A} (\vec u ; \vec x , \vec w^A) ,\Wit{\vec u ; \vec x}{B} (\vec u ; \vec x , \vec w^B)  )  \\
135 156 adas
	\smallskip
136 156 adas
	\Wit{\vec u ; \vec x}{A \cand B} (\vec u ; \vec x , \vec w^A , \vec w^B) & \dfn & \cand ( ; \Wit{\vec u ; \vec x}{A} (\vec u ; \vec x , \vec w^A) ,\Wit{\vec u ; \vec x}{B} (\vec u ; \vec x , \vec w^B)  )  \\
137 156 adas
	\smallskip
138 156 adas
	\Wit{\vec u ; \vec x}{\exists x^\safe . A(x)} (\vec u ; \vec x , \vec w , w) & \dfn & \Wit{\vec u ; \vec x , x}{A(x)} ( \vec u ; \vec x , w , \vec w )
139 156 adas
	\end{array}
140 156 adas
	\]
141 156 adas
142 156 adas
143 156 adas
\end{definition}
144 156 adas
145 156 adas
\nb{use (de)pairing to make sure only $i$ $\mu$s are used to express a $\Pi_i$ predicate.}
146 156 adas
147 156 adas
\begin{proposition}
148 156 adas
	For $\Sigma_{i+1}$ formulae $A$, $\Wit{\vec u ; \vec x}{A}$ is a $\mubci{i} $ program.
149 156 adas
\end{proposition}
150 156 adas
151 156 adas
\nb{In fact, need that $\wit{}{}$ is $\mubci{i}$ and $\Wit{}{}$ and the witness functions $\vec f$ are $\bc (\wit{}{})$}
152 156 adas
153 156 adas
Before proving the main theorem, we will need the following `witnessing lemma':
154 156 adas
155 156 adas
\begin{lemma}
156 156 adas
	If $\pvbci{i+1} $ proves a $\Sigma^\safe_{i+1}$-sequent $\Gamma \seqar \Delta$ with free variables $\vec u^\normal , \vec x^\safe$ then there are $\mubci {i}$ functions $\vec f$ such that
157 156 adas
	\[
158 156 adas
	\pvbci {i+1} \proves \Wit{\vec u , \vec x}{\bigwedge \Gamma} (\vec u ; \vec x , \vec w) \cimp \Wit{\vec u  , \vec x}{\bigvee \Delta } (\vec u ; \vec x, \vec f (\vec u ; \vec x , \vec w) )
159 156 adas
	\]
160 156 adas
	(we simply write $\Wit{\vec u ; \vec x}{A}(\vec u ; \vec x , \vec w)$, instead of $\Wit{\vec u ; \vec x}{A}(\vec u ; \vec x , \vec w) =1 $.
161 156 adas
\end{lemma}
162 156 adas
\begin{proof}
163 156 adas
	By induction on the size of a $\pvbci{i+1} $ proof.
164 156 adas
	Interesting steps below:
165 156 adas
	\begin{itemize}
166 156 adas
		\item $\neg$-right. (Can assume only applies to atomic formulae, and so no effect.)
167 156 adas
		\item $\exists$-right.
168 156 adas
		\[
169 156 adas
		\dfrac{\Gamma \seqar \Delta , A(t^\safe )}{ \Gamma \seqar \Delta, \exists x^\safe . A(x) }
170 156 adas
		\]
171 156 adas
		By the inductive hypothesis, have functions $\vec f(\vec u ; \vec x), \vec g (\vec u ; \vec x)$ such that,
172 156 adas
		\[
173 156 adas
		\pvbci{i+1} \proves
174 156 adas
		\Wit{\vec u ; \vec x}{\bigwedge \Gamma } (\vec u ; \vec x , \vec w)
175 156 adas
		\cimp
176 156 adas
		\left(
177 156 adas
		\Wit{\vec u ; \vec x}{\bigvee \Delta} (\vec u ; \vec x , \vec f (\vec u ; \vec x , \vec w) )
178 156 adas
		\cor
179 156 adas
		\Wit{\vec u ; \vec x}{A(t)} (\vec u ; \vec x , \vec g (\vec u ; \vec x , \vec w))
180 156 adas
		\right)
181 156 adas
		\]
182 156 adas
		(just use $t$ as one of the witness functions)
183 156 adas
		\item $\forall$-right. (Must be a $\Pi_i$ formula, so forget witness and compute $\wit{}{}$)
184 156 adas
		\item Contraction-right:
185 156 adas
		\[
186 156 adas
		\dfrac{\Gamma  \seqar \Delta , A ,A}{\Gamma \seqar \Delta, A}
187 156 adas
		\]
188 156 adas
		By infuctive hypothesis have functions $\vec f, \vec g^1 , \vec g^2$ such that:
189 156 adas
		\[
190 156 adas
		todo
191 156 adas
		\]
192 156 adas
		(just use conditional with a call to $\Wit{}{}$)
193 156 adas
		\item induction
194 156 adas
		\[
195 156 adas
		\dfrac{\{\Gamma , A(a) \seqar A(s_i a) , \Delta\}_{i=0,1} }{\Gamma, A(0) \seqar A(t) , \Delta}
196 156 adas
		\]
197 156 adas
	\end{itemize}
198 156 adas
\end{proof}
199 156 adas
200 156 adas
\subsection{Completeness}
201 157 adas
Here we show that every $\mubci{i}$ function is definable in $\pvbci {i+1}$:
202 156 adas
203 157 adas
\begin{theorem}
204 157 adas
	Every $\mubci i$ function is definable in $\pvbci {i+1}$.
205 157 adas
\end{theorem}
206 156 adas
207 157 adas
\anupam{This should be simple, right?}
208 157 adas
209 156 adas
\nb{WoP known as `minimization' principles in bounded arithmetic}
210 156 adas
\begin{theorem}
211 156 adas
	[Well ordering property]
212 156 adas
	\[
213 156 adas
	\pvbci{i+1} \proves \exists x^\safe . A(x) \cimp \exists  x^\safe . (A(x) \cand \forall y^\safe . (A(y) \cimp x \leq y ) )
214 156 adas
	\]
215 156 adas
\end{theorem}
216 156 adas
\begin{proof}
217 156 adas
	We work in $\pvbci{i+1}$ and show the contrapositive.
218 156 adas
	Suppose:
219 156 adas
	\begin{equation}
220 156 adas
	\label{eqn:no-least}
221 156 adas
	\forall x^\safe. (A(x) \cimp \exists y^\safe . A(y) \cand y<x )
222 156 adas
	\end{equation}
223 156 adas
	We show that,
224 156 adas
	\begin{equation}
225 156 adas
	\label{eqn:ih-wop}
226 156 adas
	\forall x. \forall y \leq a - x. (\cnot A(y) \cimp \cnot A(y + x))
227 156 adas
	\end{equation}
228 156 adas
	by polynomial induction on $x$.
229 156 adas
230 156 adas
	Let $B(x)$ be such that \eqref{eqn:ih-wop} is $\forall x . B(x)$.
231 156 adas
	\nb{If $A \in \Sigma^\safe_i \cup \Pi^\safe_i$ then $B \in \Pi^\safe_{i+1}$.}
232 156 adas
233 156 adas
	When $x=0$, notice that \eqref{eqn:ih-wop} is just a generalised identity.
234 156 adas
235 156 adas
	Suppose that $B(x)$ and let us show that $B(2x)$.
236 156 adas
	Let $y \leq a - 2x$ such that $\cnot A(y)$.
237 156 adas
	Then $y\leq a-x$ so by $B(x)$ we have that $\cnot A(y+x)$.
238 156 adas
	We also have that $y+x \leq a-x$ so by $B(x)$ we have that $\cnot A(y+2x)$, as required.
239 156 adas
240 156 adas
	Now suppose that $B(x)$ and let us show that $B(2x+1)$.
241 156 adas
	Let $y \leq a - 2x - 1$ such that $\cnot A(y)$.
242 156 adas
	By similar reasoning to the $2x$ case, we have that $\cnot A(y + 2x )$.
243 156 adas
\end{proof}
244 156 adas
245 156 adas
246 157 adas
\begin{theorem}
247 157 adas
248 157 adas
\end{theorem}
249 157 adas
250 157 adas
251 156 adas
\subsection{What we want for WoP}
252 156 adas
From bounded arithmetic:
253 156 adas
254 156 adas
$\Sigma_{i+1}$-LMIN $ \iff$ $\Sigma_{i+1}$-PIND $\implies$ $\Sigma_i$-IND $\iff$ $\Sigma_i$-MIN $ \iff$ $\Pi_{i+1}$-MIN.
255 156 adas
256 156 adas
\subsection{Completeness proof idea}
257 156 adas
For each $\mubci i$ function $f(\vec u ; \vec x)$ we $\Sigma_i$-define a formula $A_f (\vec u ; \vec x  , y )$ in $\pvbci{i+1}$ such that:
258 156 adas
\[
259 156 adas
\proves A_f (\vec u ; \vec x , y)
260 156 adas
\quad
261 156 adas
\iff
262 156 adas
\quad
263 156 adas
f(\vec u ; \vec x) = y
264 156 adas
\]
265 156 adas
and $A_f$ is provably total in $\pvbci{i+1}$.
266 156 adas
267 156 adas
For the $\mu$ case, say we have the function:
268 156 adas
\[
269 156 adas
\mu x^{+1} . f(\vec u ; \vec x , x) =_2 0
270 156 adas
\]
271 156 adas
Let $A_f (\vec u ; \vec x , y)$ be given by the inductive hypothesis.
272 156 adas
We define $A(\vec u ; \vec x , z)$ as:
273 156 adas
\[
274 156 adas
\begin{array}{rl}
275 156 adas
&\left(
276 156 adas
z=0 \  \cand \ \forall x^\safe , y^\safe . (A_f (\vec u ; \vec x , x, y) \cimp y=_2 1)
277 156 adas
\right) \\
278 156 adas
\cor & \left(
279 156 adas
\begin{array}{ll}
280 156 adas
z\neq 0
281 156 adas
& \cand\   \forall y^\safe . (A_f (\vec u ; \vec x , z , y) \cimp y=_2 0 ) \\
282 156 adas
& \cand\ \forall x^\safe < p(;z) . (\forall y^\safe . A_f (\vec u ; \vec x , x , y) \cimp y=_2 1)
283 156 adas
\end{array}
284 156 adas
\right)
285 156 adas
\end{array}
286 156 adas
\]
287 156 adas
Notice that $A$ is $\Pi_k$, since $A_f$ is $\Sigma_k$.
288 156 adas
289 156 adas
290 156 adas
291 156 adas
What about, say recursion on a formula? Need a form of `ranked comprehension'?
292 156 adas
E.g., when $A$ is $\Sigma_k$ then we can introduce a rank $k$ symbol (a sort?) such that:
293 156 adas
\[
294 156 adas
\forall \vec u^\normal, \vec x^\safe . \exists ! y^\safe . A(\vec u ; \vec x , y)
295 156 adas
\implies
296 156 adas
\exists f^\safe_r . \forall \vec u^\normal,\vec x^\safe, y^\safe . (A(\vec u ; \vec x, y) \ciff f^\safe_r (\vec u ; \vec x) = y )
297 156 adas
\]
298 156 adas
299 156 adas
Otherwise, can we use definability of computations? E.g., if:
300 156 adas
\[
301 156 adas
\begin{array}{rcl}
302 156 adas
f(0, \vec u ; \vec x ) & \dfn & g(\vec u ; \vec x) \\
303 156 adas
f(s_i u , \vec u ; \vec x) & \dfn & h_i (u , \vec u ; \vec x , f(u,\vec u ; \vec x))
304 156 adas
\end{array}
305 156 adas
\]
306 156 adas
Suppose we have $A_g (\vec u ; \vec x,y)$ and $A_i (u , \vec u ; \vec x , y , z)$ defining $g$ and $h_i$ respectively.
307 156 adas
We define $A_f (u ,\vec u ; \vec x , y)$ as:
308 156 adas
\[
309 156 adas
\exists z^\safe . \left(
310 156 adas
\begin{array}{ll}
311 156 adas
& Seq(z) \cand \exists y_0 . ( A_g (\vec u ; \vec x , y_0) \cand \beta_0 (z , y_0) ) \cand \beta_{|u|} ( z,y ) \\
312 156 adas
\cand & \forall k < |u| . \exists y_k , y_{k+1} . ( \beta_k (z, y_i) \cand \beta_{k+1} (z, y_{k+1})  \cand A_i (u , \vec u ; \vec x , y_k , y_{k+1}) )
313 156 adas
\end{array}
314 156 adas
\right)
315 156 adas
\]
316 156 adas
317 156 adas
(Can we really assume $z$ is safe here?)
318 156 adas
319 156 adas
320 156 adas
POINT: for whatever formulation, we need to prove:
321 156 adas
\[
322 156 adas
\exists y^\safe . A_f (a , \vec u ; \vec x , y)
323 156 adas
\quad \seqar \quad
324 156 adas
\exists y^\safe . A_f (s_i a, \vec u ; \vec x , y)
325 156 adas
\]
326 156 adas
327 156 adas
SHOULD HAVE: $\beta (i;x)$ for $i$th element of sequence $x$.
328 156 adas
Therefore need 'sharply bounded' quantification for normal variables?
329 156 adas
330 156 adas
In fact, why not $\beta(;i,x) $? Should be fine. So only safe quantification is needed for PH, but lose level-by-level delineation.
331 156 adas
332 156 adas
GOALS:
333 156 adas
\begin{enumerate}
334 156 adas
	\item PVBC + FCA + $\safe$-IND characterises PH. (Recursion included in PVBC)
335 156 adas
	\item Refinement of above with `ranks' to delineate levels (definitions of $\pvbci{i}$).
336 156 adas
	\item Arithmetic including both safe and sharply bounded normal quantification. (for sequences)
337 156 adas
	\item (if time) allow both bounded and safe quantifiers?
338 156 adas
\end{enumerate}
339 156 adas
340 156 adas
341 156 adas
342 156 adas
FCA:
343 156 adas
\[
344 156 adas
\exists f^\safe . \forall \vec u ; \vec x .
345 156 adas
\left(
346 156 adas
\exists y^\safe . A(\vec u ; \vec x , y) \ciff A(\vec u ; \vec x , f^\safe(\vec u ; \vec x))
347 156 adas
\right)
348 156 adas
\]
349 156 adas
(with typing information)
350 156 adas
This could be enough with open induction if we introduce ranks later? Yup, seems like a good idea. Can then make into a real `open' theory.
351 156 adas
352 157 adas
\subsection{Delineating levels using function ranks}
353 157 adas
354 157 adas
355 157 adas
\subsection{A second-order theory}
356 157 adas
From here we can rather simply define a theory complete for PH.
357 157 adas
However, delineating the levels is a little cumbersome, due to the formal necessity of infinitely many ranks.
358 157 adas
359 157 adas
\begin{definition}
360 157 adas
	[Axioms]
361 157 adas
	The \emph{functional comprehension} schema is the following:
362 157 adas
	\[
363 157 adas
	\exists f . \forall \vec u; \vec x . ( \exists y^\safe . A(\vec u ; \vec x , y) \ciff A(\vec u ; \vec x , f(\vec u ; \vec x) )
364 157 adas
	\]
365 157 adas
	(can parametrise by which $A$ permitted)
366 157 adas
367 157 adas
	The \emph{recursion} schema is:
368 157 adas
	\[
369 157 adas
	\forall g , h_0 , h_1 . \exists f . \forall u , \vec u ; \vec x .
370 157 adas
	\left(
371 157 adas
	\begin{array}{rl}
372 157 adas
	& f(0 , \vec u ; \vec x) = g(\vec u ; \vec x) \\
373 157 adas
	\cand & f(\succ 0 u , \vec u ; \vec x) = h_0 (u , \vec u ; \vec x , f(u , \vec u ; \vec x)) \\
374 157 adas
	\cand & f(\succ 1 u , \vec u ; \vec x) = h_1 (u , \vec u ; \vec x , f(u , \vec u ; \vec x))
375 157 adas
	\end{array}
376 157 adas
	\right)
377 157 adas
	\]
378 157 adas
	(should be well typed. Cannot avoid due to sequentiality problem.)
379 157 adas
	\end{definition}
380 157 adas
381 157 adas
	\anupam{Original PV has explicit recursion symbols. Also, Krajicek's PVi has explicit symbols for the characteristic functions of each $\Sigma^b_i$ predicate.}