Statistiques
| Révision :

root / CSL17 / tech-report / pv-theories.tex @ 253

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

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