pith. sign in

arxiv: 2605.17801 · v1 · pith:6SCQKSQZnew · submitted 2026-05-18 · 💻 cs.IT · cs.NA· math.IT· math.NA· nlin.CG

The information-theoretic complexity of differentiable functions

Pith reviewed 2026-05-20 01:35 UTC · model grok-4.3

classification 💻 cs.IT cs.NAmath.ITmath.NAnlin.CG
keywords V-complexityeffective complexitydifferentiable functionpiecewise constant approximationinformation theoretic complexitycomplex systemsdiffusion
0
0 comments X

The pith

V-complexity quantifies the complexity of differentiable functions using piecewise constant approximations.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper defines a complexity measure for differentiable functions called V-complexity. It is based on the quality of approximation by piecewise constant functions along with the number of intervals required. This approach is intended to capture intuitive notions of simplicity and complexity in functions. The measure is conjectured to match the compressibility of the function under run length encoding and Lempel-Ziv 77. When incorporated into the definition of effective complexity for complex systems, it produces a rise and subsequent fall in complexity for the diffusion of cream in coffee, matching other methods.

Core claim

V-complexity is introduced as a measure that takes into account the quality of the approximation and the number of intervals in the approximating function. It is shown to formalize some intuitions about the simplicity or complexity of f(x). The V-complexity is hypothesized to be equivalent to the compression measure for the Run Length Encoding and the Lempel Ziv 77 algorithms. This allows the effective complexity of a complex system to be defined as the V-complexity of the differentiable function describing its perceived regularities, as demonstrated in the diffusion of cream in a cup of coffee where it starts at zero, increases to a maximum, and decreases back to zero.

What carries the argument

V-complexity, a measure combining the error in approximating a differentiable function by a piecewise constant function with the number of intervals used in that approximation.

If this is right

  • V-complexity can be used to define the effective complexity of complex systems when their regularities are given by a differentiable function.
  • In the cream diffusion model, V-complexity starts at zero, rises quickly to a maximum, then falls back to zero at equilibrium.
  • The same rise and fall behavior is obtained using a cellular automaton and the concept of apparent complexity.
  • V-complexity is equivalent to the compression measure using run length encoding and Lempel Ziv 77.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • This measure provides a way to track the evolution of complexity in physical systems over time without relying on specific simulation methods.
  • It suggests that many systems may exhibit maximum complexity during their transition to equilibrium rather than at the start or end.
  • The hypothesized equivalence to compression could allow using V-complexity as a faster proxy for compressibility in function data.

Load-bearing premise

The premise that approximations by piecewise constant functions, weighted by both their accuracy and the number of pieces, provide a general and meaningful information-theoretic complexity measure for differentiable functions.

What would settle it

A specific differentiable function for which V-complexity gives a value that contradicts the compressibility under run length encoding or Lempel-Ziv 77, or that does not align with standard intuitions about its complexity.

Figures

Figures reproduced from arXiv: 2605.17801 by Matthijs Ruijgrok.

Figure 1
Figure 1. Figure 1: Some functions with domain and range equal to [0 [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The step function approximation for f(x) = x 2 , with ε = 0.07. In the following list, V (f) is calculated for some (families of) functions which demon￾strate essential properties of this complexity measure. (a) Quadratic function f(x) = x 2 on [−1, 1]. From (7) it follows that V (f) = 1 4 Z 1 −1 |2x| 1 2 dx 2 = 8 9 . 6 [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Complexity of fα(x) = x α. On the horizontal axis are the values of 0 ≤ α ≤ 5. Red dots indicate the RLE-complexity C(fα), the blue graph shows V (fα). assuming the limit exists. As with V-complexity, the RLE-complexity is the product of the amount of information in the RLE-approximation with the error of the approx￾imation, as the error goes to zero. However, these two ingredients depend on the grid ratio… view at source ↗
Figure 4
Figure 4. Figure 4: Comparison of RLE, GZIP and Compress encoding of [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Ratio of the encoding length of f(x) = sin2 πx on the interval [0, 1] to encoding length over [0, 1/2]. Horizontally are the number of grid points. M = 4096 and r = 0.5. respectively. This is repeated for values of n ranging from n = 100 to n = 104 . In figure 5b, the ratios l2/l1 are plotted. For n < M, this ratio is close to 1. This is because the LZ77 algorithm uses the fact that f(x), restricted to [1/… view at source ↗
Figure 6
Figure 6. Figure 6: The lattice gas model. Dashed lines represent coffee particles, solid [PITH_FULL_IMAGE:figures/full_fig_p012_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: State of the automaton at four different times. Parameters are [PITH_FULL_IMAGE:figures/full_fig_p014_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Fraction of cream particles in upper half of the cup. [PITH_FULL_IMAGE:figures/full_fig_p015_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Apparent complexity of lattice gas automaton. The compression [PITH_FULL_IMAGE:figures/full_fig_p015_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: V-complexity of the solution of (14) (solid line) and RLE-complexity [PITH_FULL_IMAGE:figures/full_fig_p016_10.png] view at source ↗
read the original abstract

A measure for the complexity of a differentiable function f(x) on an interval is introduced. It is based on approximations of the function by piecewise constant functions. The measure takes into account the quality of the approximation and the number of intervals in the approximating function. This measure, called the V-complexity of f(x), is shown to formalize some intuitions about the simplicity or complexity of f(x). The V-complexity is then compared to another measure of complexity, namely how compressible an approximation of f(x) is. It is hypothesized that V-complexity is equivalent to the compression measure, in the case of the Run Length Encoding and the Lempel Ziv 77 algorithms. V-complexity can be used as an ingredient in the definition of the Effective Complexity (EC) of a Complex System. When the perceived regularities of such a system are described by a differentiable function on an interval, the EC can be defined as the V-complexity of that function. EC is applied to the model of diffusion of cream in a cup of coffee. The perceived regularity of this model is given by the diffusion equation. The V-complexity of the solution of the equation starts at zero, quickly increases to a maximum and then decreases back to zero as the liquid reaches its equilibrium state. It is shown that this is also the result when a cellular automaton approach and the concept of Apparent Complexity is used.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript introduces V-complexity, a measure for the complexity of a differentiable function f(x) on an interval. It is defined via the quality of approximation by piecewise-constant functions together with the number of intervals required. The paper claims that this measure formalizes intuitions about simplicity and complexity of functions. It hypothesizes equivalence between V-complexity and compressibility under Run-Length Encoding and Lempel-Ziv 77. V-complexity is then proposed as the basis for Effective Complexity (EC) of a complex system whose perceived regularities are captured by such a function; the diffusion equation solution for cream mixing in coffee is used as an example, with V-complexity shown to rise from zero to a peak and return to zero, matching the behavior obtained from a cellular-automaton model and Apparent Complexity.

Significance. If the hypothesized link to compression algorithms can be made rigorous and the measure shown to be independent of the specific approximation scheme, V-complexity would supply a concrete, approximation-based route to information-theoretic complexity for differentiable functions and a practical ingredient for Effective Complexity. The diffusion example, together with its consistency check against cellular automata, illustrates how the measure can track the emergence and disappearance of regularities in a physical model; this is a concrete strength that could be developed further.

major comments (2)
  1. [Abstract / hypothesis paragraph] Abstract (paragraph introducing the hypothesis): the claim that V-complexity is equivalent to the compression length under RLE and LZ77 is stated as a hypothesis, yet no explicit construction, choice of approximation rule (uniform grid, adaptive, or minimax), error metric, or side-by-side numerical comparison is supplied that would show the resulting V-value equals or ranks identically with the compressed bit length of the same discretized function. Because this equivalence is the stated bridge to information theory, its absence leaves the central interpretive claim unanchored.
  2. [Definition of V-complexity] Definition of V-complexity (section following the abstract): the quality of piecewise-constant approximation and the interval count are asserted to yield a meaningful complexity measure that captures perceived regularities independently of the concrete approximation algorithm or error metric. No robustness check or invariance argument is given to support this independence, which is load-bearing for the claim that V-complexity is a general information-theoretic quantity.
minor comments (1)
  1. [Diffusion application] The diffusion example would benefit from an explicit statement of the discretization used to compute V-complexity from the analytic solution of the diffusion equation.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We agree that the hypothesis linking V-complexity to compression requires concrete support and that independence from approximation details needs explicit justification. Below we respond point by point and indicate the planned revisions.

read point-by-point responses
  1. Referee: [Abstract / hypothesis paragraph] Abstract (paragraph introducing the hypothesis): the claim that V-complexity is equivalent to the compression length under RLE and LZ77 is stated as a hypothesis, yet no explicit construction, choice of approximation rule (uniform grid, adaptive, or minimax), error metric, or side-by-side numerical comparison is supplied that would show the resulting V-value equals or ranks identically with the compressed bit length of the same discretized function. Because this equivalence is the stated bridge to information theory, its absence leaves the central interpretive claim unanchored.

    Authors: We accept the observation. The equivalence is presented as a hypothesis without numerical validation or specification of the approximation procedure. In the revision we will add a new subsection that fixes the approximation rule to uniform partitioning with supremum-norm error, supplies explicit constructions for several test functions, and reports side-by-side comparisons of the resulting V-values against the compressed lengths produced by RLE and LZ77 on the same discretizations. This will provide the missing anchor for the information-theoretic interpretation. revision: yes

  2. Referee: [Definition of V-complexity] Definition of V-complexity (section following the abstract): the quality of piecewise-constant approximation and the interval count are asserted to yield a meaningful complexity measure that captures perceived regularities independently of the concrete approximation algorithm or error metric. No robustness check or invariance argument is given to support this independence, which is load-bearing for the claim that V-complexity is a general information-theoretic quantity.

    Authors: We agree that an invariance argument or robustness demonstration is required. We will revise the definition section to include a short asymptotic argument showing that, for fine enough partitions, the V-complexity ordering is independent of the choice between uniform and adaptive grids and between L2 and L-infinity error metrics. We will also add a brief numerical check on a small collection of example functions confirming that relative complexities remain stable across these variants. revision: yes

Circularity Check

0 steps flagged

V-complexity defined from first principles via piecewise-constant approximation; no reduction to inputs or self-citation chain.

full rationale

The paper introduces V-complexity directly from the quality of approximation by piecewise constant functions together with the number of intervals; this construction is independent of any compression length or fitted parameter. The subsequent hypothesis of equivalence to RLE and LZ77 is stated as a conjecture without any equation that equates the two measures by definition or by construction. The application to effective complexity of the diffusion equation and the matching result with cellular automata is presented as an external consistency check rather than a self-referential derivation. No self-citation load-bearing step, ansatz smuggling, or renaming of a known result appears in the derivation chain. The measure therefore remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The central claim rests on the new definition of V-complexity and the unproven hypothesis of equivalence to specific compression algorithms. No numerical free parameters are stated. The primary invented entity is the V-complexity measure itself.

axioms (1)
  • domain assumption Differentiable functions on an interval admit useful approximations by piecewise constant functions whose quality and piece count can serve as a complexity measure.
    Invoked at the start of the V-complexity definition in the abstract.
invented entities (1)
  • V-complexity no independent evidence
    purpose: Quantify the information-theoretic complexity of a differentiable function via piecewise-constant approximation quality and interval count.
    Newly defined measure with no external validation or independent evidence supplied in the abstract.

pith-pipeline@v0.9.0 · 5786 in / 1491 out tokens · 58656 ms · 2026-05-20T01:35:12.653122+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

15 extracted references · 15 canonical work pages

  1. [1]

    Royden and P.M

    H.L. Royden and P.M. Fitzpatrick. Real Analysis, 5th edition. 2022

  2. [2]

    Huang and R.D

    W. Huang and R.D. Russel. Adaptive moving mesh methods. 2010

  3. [3]

    Chopard and M

    B. Chopard and M. Droz. Cellular Automata Modeling of Physical Systems. 2005

  4. [4]

    European Journal for Philosophy of Science

    What is a complex system , author =. European Journal for Philosophy of Science. 2012. doi:10.1007/s13194-012-0056-8

  5. [5]

    Journal of Computational Physics

    A robust moving mesh finite volume method applied to 1D hyperbolic conservation laws from magnetohydrodynamics , author =. Journal of Computational Physics. 2006. doi:10.1016/j.jcp.2005.12.014

  6. [6]

    IMA Journal of Numerical Analysis

    Uniform convergence analysis of an upwind finite-difference approximation of a convection–diffusion boundary value problem on an adaptive grid , author =. IMA Journal of Numerical Analysis. 1999. doi:10.1093/imanum/19.2.233

  7. [7]

    Journal of Computational and Applied Mathematics

    Analysis of difference approximations to a singularly perturbed two-point boundary value problem on an adaptively generated grid , author =. Journal of Computational and Applied Mathematics. 1999. doi:10.1016/S0377-0427(98)00136-8

  8. [8]

    American Journal of Physics

    Resource Letter CS-1: Complex Systems , author =. American Journal of Physics. 2012. doi:10.1119/1.3590372

  9. [9]

    Complexity

    An Information-Theoretic Primer on Complexity, Self-Organization, and Emergence , author =. Complexity. 2009. doi:10.1002/cplx.20249

  10. [10]

    2014 , eprint=

    Quantifying the Rise and Fall of Complexity in Closed Systems: The Coffee Automaton , author=. 2014 , eprint=

  11. [11]

    URLhttps://doi.org/10.1038/nn831

    Efficient coding of natural sounds , author =. Nature Neuroscience. 2002. doi:10.1038/nn831

  12. [12]

    Real Diffusion Experiment (for Home or School)

    Nir Livne. Real Diffusion Experiment (for Home or School)

  13. [13]

    Sean M. Carroll. 15th annual Biard lecture at Ohio State University

  14. [14]

    and Lempel, A

    Ziv, J. and Lempel, A. , journal=. A universal algorithm for sequential data compression , year=

  15. [15]

    Complexity , volume =

    Gell-Mann, Murray and Lloyd, Seth , title =. Complexity , volume =. doi:https://doi.org/10.1002/(SICI)1099-0526(199609/10)2:1<44::AID-CPLX10>3.0.CO;2-X , year =