Abstract
In this paper we show that several classes of languages from computational complexity theory, such as
Keywords
Introduction
In the papers [3,5] the authors presented a characterization of the computational complexity class P by using polynomial ordinary differential equations (ODEs). This characterization is purely continuous and hence establishes a continuous characterization of the class P typically associated to discrete models of computation such as the standard Turing machine. This result does not use any reference to a (discrete) machine and hence provides an implicit characterization of P.
Interestingly, this result also provides a form of equivalence between analog and digital models of computations, both at a computability level and at a complexity level. This stems from the fact that the class of functions which can be computed with these polynomial ODEs corresponds to an analog model of computation, Shannon’s General Purpose Analog Computer (GPAC) [18,24], which is intended to model differential analyzers, which were the analog computers in use before the advent of the digital computer [11]. It was shown that the GPAC and Turing based models of computation, including computable analysis, are computationally equivalent at a computability level [1,2,17] and at a complexity level, when we consider functions computable in polynomial time [3,5].
A natural question to ask is whether the above continuous characterization of polynomial time computability via ODEs extends to other computational complexity classes. This is not obvious from [3,5], since these papers are technically dense, and rely on other papers such as [4] which are tailored to the polynomial case. Polynomials have many desirable properties which are unfortunately not shared by some other types of functions. For example, the class of polynomials is closed under composition, while the class of exponential functions is not, which may (and will, as we will see later) pose problems when trying to characterize the class
The purpose of the present paper is to understand how the results of [3,5] can be extended to other complexity classes such as
As we will see in the following sections, we will be able to reuse some results from previous papers such as [3–5], while other results will have to be adapted. As we mentioned, the papers [3,5] are technically dense and often not easy to follow. In this paper we also intend to provide a “road map” to the results of [5], which might help simplify their understanding, by clearly separating the “core” results of the papers [3,5], from their “non-core” counterparts.
To obtain our results, first we characterize the class
Finally, combining our approach with that of [5] we show that we can characterize classes of languages instead of classes of functions, and this will be concretely implemented for describing the complexity class
The outline of the paper is as follows. On Section 2 we review some basic notions about computability with ODEs. In Section 3 we briefly review the results of [5] which show how one can characterize the class
Computing with polynomial differential equations
As we have mentioned, polynomial ODEs correspond to Shannon’s GPAC. More concretely a function is generable if it is the solution of some polynomial initial-value problem (PIVP) defined with a polynomial ODE. We call such functions PIVP functions. Note that in this sense a generable function is a one-variable function. The notion of generable function can be extended to multivariate functions as in [5]. Moreover, by restricting the coefficients of the polynomials used to define the ODE, we can also restrict the class of functions computed by this model. This is usually assumed to avoid potential problems where the ODE gains unreasonable computational power by using the coefficients of the polynomials as oracles. More concretely, let
Let
We remark that, for the class
We say that
If
We note that several standard functions from analysis such as the trigonometric functions, polynomials, the exponential and logarithmic functions all belong to
In what follows, it will be convenient to consider generable functions which are bounded by a polynomial. We say that

Computing the function f with an ODE.
Let
The function Π is usually referred to as the time bound, and the function Υ as the space bound (the notation
Let
When using the class
Let
For any
The first condition of Proposition 6 defines the ODEs generating the function f, the second condition ensures that the norm of the solution of this system satisfies a polynomial bound, while the third condition takes care of the convergence of such solution: specifically, this solution y converges to the correct value of the function
The class
If
(ATSP modulus of continuity).
If
One the most important properties of
(ATSP closure by composition).
If
Since the class
A set X is called a star domain if there exists
The following theorem is from [5].
If
One key element to extend the polynomial characterization of P and
In this section we review several useful results from [5]. In particular we will see how we can code the transition function of a Turing machine (TM for short) into a function in
Simulating Turing machines
In this section we will encode the configuration of a Turing machine M as an element of
Let
The real functions
Other two crucial functions are

The graphs of functions
For every
If
If
In all cases,
Before showing how a configuration can be encoded into an element of
(Turing Machine).
A Turing Machine is a tuple
Note that each symbol
(Configuration).
A configuration for a Turing Machine M is a tuple
We now show how a configuration can be encoded as an element of
Let
Let M also denote the transition function of the Turing machine M, i.e. if c is a configuration, then
Let M be a Turing machine with an alphabet
To simulate the computation of a Turing machine M over the real numbers, we need not only to be able to code the transition function of M as a function over the reals, but also to iterate this function over the set of (the coding of) its configurations
The following theorem is from [5] and shows that a function f in
([5], Polynomial iteration of a function in ATSP).
Let
For all
For all
Let
Equivalence relation between
and
Using the results of the previous section it can be shown [5] that
(Discrete emulation).
Let G be a set of functions from
Note that the above definition could be generalized to consider other encodings from
As an intuition behind the above definition, a function g defined over a continuous domain emulates a function f defined over a discrete domain if the operation of encoding commutes with the applications of the functions. This means that, given a fixed input word from a discrete alphabet, encoding from the discrete to the continuous can be done either directly to the input before the application of function g, either after the application of function f to the input, yielding the same result.
For simplicity, later on we will use the notation
([5], FP equivalence).
Let
To show that if
In [5], it was shown that there is a function
Computing discrete functions: The exponential case
One might think that the class
In the present paper we solve the above problem and define a class
We say that a function
We first define the class
Our main insight to solve the above problem is that, contrarily to what was done for the case of
Another problem that we face is that we expect
(ATSE).
Let
In other words, a function belongs to
(GEVAL).
Let D be a domain in
Note that, since exponential functions can be easily expressed as solutions of polynomial ODEs (for example
Properties of the exponential analog classes
Using straightforward adaptations of the proofs presented for their polynomial counterparts in [5], we get the following properties:
If
As a consequence of the above theorem together with our previous observation, exponential functions defined over star domains belong to
Let
As mentioned earlier, the closure of the
If
We present the proof only for the closure by product, since the other cases are similar. We prove the theorem for dimension one, but the proof can be easily repeated for higher dimensions. Let us consider a function
Let f be a function,
Let Let us now tackle the time bound of Definition 23. Define (ATSE closure by arithmetic operations).
Finally, we can state one of the main results of this paper. By adapting the construction already developed in [5], we can show the equivalence between the class
(FEXPTIME equivalence).
Let
Next we proceed with the proof of Theorem 29. We will prove the direct and reverse direction of the theorem separately.
For the direct direction of Theorem 29, we have to show that if
Let us assume that
Let
In particular, we get that
Recalling that
We now proceed with the reverse direction of the proof. In other words, we have to show that if f is emulable under
([23]).
Let
More precisely, the theorem states that there is a Turing machine M such that for any oracle O representing
Now we can continue with the proof of the reverse direction of Theorem 29. Assume that f is emulable under
Let
Let
Going beyond the exponential case
As we have showed previously in this paper, the key intuition of splitting the dependence of the time and space boundaries from one single term, Υ, into the product of two separate components
Following every step of the proofs, starting from the basic properties and definitions of the
In this section we extend the result of Theorem 21 to other complexity classes. More precisely, we list which conditions the analog classes involved have to satisfy in order to repeat the simulation process already obtained for the polynomial case. First, let us define a version of the
Let
As we already discussed in the previous section, we need to be able to enforce closure by composition and by arithmetic operations for the class
Let
We now present conditions that ensure that the result of Theorem 21 can be extended to other complexity classes.
(Sufficient conditions).
Let A be a class of functions from If If p is polynomial and If If
The first condition enforces a form of closure for the main arithmetical operations which are of interest to us. The second condition provides enough elements so that we can have a variant of Theorem 28 for the class
Note that
We recall that if
We note that, similarly to what happens to
At this point we can finally state the following generalized equivalence theorem.
Let A be a class of functions that satisfies the conditions of Definition
33
. Then given a function
We will use this theorem to characterize the Grzegorczyk hierarchy with ODEs. In particular, we will also be able to characterize the class of elementary computable functions and the class of primitive recursive functions with ODEs. Nonetheless, before proceeding with the application of Theorem 35 above to the case of the Grzegorczyk hierarchy, in the next section we show that, similarly to the case of the polynomial class
Generalization of the Analog Length class
We start this section by introducing the definitions required to understand the meaning of the Analog Length class.
Let
In Definition 36 the first item has the same role of the first condition of the definition of
Let
Notice that Definition 37 is just the generalization of Definition 36 when the boundary Π is taken from a set A, exactly as definition of
To proceed with the arguments of this section we now need to introduce a lemma showing some useful boundaries over multivariate polynomials.
Let
We have
Let us now assume that the set A considered in this section satisfies the conditions listed above in Definition 33. Then we obtain an equivalence between the two analog classes in the form of the following theorem: If A is a class of functions which satisfies the conditions of Definition
33
, then The proof of this theorem follows essentially the structure of the proof of Theorem 18 in [4]. Let us first suppose that We begin by noting that if For the reverse implication, let us suppose that Let us thus assume that This shows that
We start this section by briefly recalling the definition of the Grzegorczyk hierarchy. The Grzegorczyk hierarchy, originally proposed by Andrzej Gregorczyk in 1953 in [19], is a hierarchy of classes of functions from the naturals to the naturals, defined recursively. For our specific purpose, the first two levels of the hierarchy are not relevant, since they only include trivial functions such as addition and multiplication, which are obviously computable in polynomial time. The third level, which we indicate with the notation
(Grzegorczyk Hierarchy).
For
It can be shown that each level is properly included in the next one,
Analog characterization of each level
In this section our objective is to use ODEs to characterize the classes
However, a problem arises if one wants to use Theorem 35 to characterize
With this objective in mind, we now present a construction to iterate functions with ODEs which is based on the ODE
Let us first assume that If
One can also consider the more general targeting ODE
We also note (to our knowledge, this was not mentioned in the literature before) that c might not be a fixed value, but instead some function
We will now show how the targeting ODE (17) can be used to obtain a function

Let
We can see the function
Proceeding again as in [17], let us take the function s defined by

The graph of the function s defined on equation (18).
The following function σ is also from [17, Proposition 5].
Let
The function σ behaves as a uniform contraction in a neighborhood of the integers
Let us now show how we can obtain a function

Iterating a function with an ODE. Here the dashed (red) and blue (solid) graphs represent the ideal behavior of
Using a similar argument we conclude that
Let us now analyze in more detail the ODE (20). When
This result can be generalized as shown in the following theorem.
Given the function
The proof is done by induction on n. The base case was performed just before the proof of this theorem (note that
Now we go to the induction step. Suppose that Iterating the function 
We note also that, by construction
On the first half-unit interval, where
We now proceed with the second time interval. Since
Now the procedure repeats itself on the time interval
By repeating this procedure on subsequent intervals and by considering w as the variable
We now know from the previous theorem that each function
For each
Note that this definition of
It is then left to prove condition 3, which requires that given any function
From the above argument, we have just showed by means of the above proof of Theorem 43 that each
To show condition (i), we first present the two following lemmas taken from [6, Lemma 24, Corollary 26]:
Let
([6], Generable functions are closed under ODE).
Let
Then
Since Lemmas 46 and 47 show closure of
First, define
To show that
Let us now assume that
At this point we have shown that
If
We note that (necessarily univariate) functions of
Let
This lemma tells us that each function in
Let
Therefore, according to Definition 31, we have just shown that the second condition of this definition is satisfied when using (23) to approach
By Lemma 49, we know that there exists
Hence, the classes
Let
Characterizing the class
Until now we have described connections between analog classes of functions (e.g.
Definitions of the analog classes
We start with a definition of a new class of sets, that we will call Exponential-Analog-Recognizable, or
(EAR).
A language if if

Graphical depiction of Definition 51. A word w is accepted (rejected) if after following the solution curve for a length exponential in
Informally, the definition above states that a language L belong to the
We can now state the equivalence result related to the
For any language
Let From the definition of the Let us now consider the case when Note that We will now proceed with the reverse direction of the proof of Theorem 52. Assume that At this point some extra care is necessary. Indeed, the temptation is to use Theorem 30 to compute with a desired precision the value of the curve (EXPTIME equivalence).
In this paper we have showed that the analog characterization obtained in [5] for polynomial complexity classes can be applied to other classes of computable functions. To reach this conclusion we had to overcome the fact that the equivalence established at a polynomial level was tailored over properties of polynomials, and therefore not trivially applicable to more general classes of functions. We started by analyzing the exponential case and we proved that the key ingredient to obtain the analog characterization was to modify the analog classes involved in such a way that could ensure closure by basic arithmetic operations and closure by composition with functions in
As a natural consequence of the work presented with this paper, one can wonder whether a similar characterization can be obtained for complexity classes such as
Footnotes
Acknowledgements
The authors wish to thank Olivier Bournez and Amaury Pouly for helpful discussions. R. Gozzi was supported by the Fundação para a Ciência e a Tecnologia (FCT) via grant PD/BD/135190/2017. This work is funded by FCT/MCTES through national funds and when applicable co-funded EU funds under the project UIDB/50008/2020.
