pith. sign in

arxiv: 2509.06238 · v2 · submitted 2025-09-07 · 🧮 math.LO

On Approximate Classification of Theories

Alexander Burka This is my paper

Pith reviewed 2026-05-18 17:36 UTC · model grok-4.3

classification 🧮 math.LO
keywords approximate logicmodel theorystabilitysimplicityfirst-order logictheory classificationlogic
0
0 comments X

The pith

Stability and simplicity extend to approximate first-order logic while preserving key classical results.

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

The paper sets out a framework that carries the model-theoretic ideas of stability and simplicity over to an approximate version of first-order logic. In this version, the truth of formulas is allowed to hold to varying degrees rather than exactly. A reader would care if the framework lets the same dividing lines and classification tools apply to logics that better match noisy or graded data. The central move is to show that several standard theorems about when a theory is stable or simple still go through once the approximate setting is properly defined.

Core claim

We propose a framework for model-theoretic stability and simplicity in an approximate first-order setting and generalize some classical results.

What carries the argument

The approximate first-order logic framework, which redefines satisfaction relations to allow graded truth while keeping stability and simplicity definable.

If this is right

  • Theorems that characterize stable theories in the exact case have counterparts in the approximate case.
  • Simplicity can be detected by the same non-dividing-line criteria once approximation is built in.
  • The overall classification of theories by their model-theoretic properties continues to function.
  • Approximate structures become classifiable using the same dividing lines as exact structures.

Where Pith is reading between the lines

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

  • The framework could be tested on concrete approximate structures such as noisy graphs or metric spaces to see which classical invariants survive.
  • If the generalizations hold, similar approximate versions might be developed for other dividing lines such as NIP or o-minimality.
  • Connections may appear between this setting and existing work on continuous or fuzzy logics.

Load-bearing premise

An approximate first-order logic can be set up so that the classical definitions of stability and simplicity remain meaningful and the usual theorems still hold.

What would settle it

A concrete approximate logic in which one of the generalized stability or simplicity theorems fails to hold, or in which no consistent dividing line can be defined.

read the original abstract

We propose a framework for model-theoretic stability and simplicity in an approximate first-order setting and generalize some classical results.

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

0 major / 3 minor

Summary. The manuscript proposes a framework for model-theoretic stability and simplicity in an approximate first-order setting. It defines approximate formulas and types via a metric or tolerance relation on the language, re-derives forking and dividing in this context, and proves that classical equivalences (stability equivalent to absence of the order property; simplicity equivalent to absence of the tree property of the second kind) continue to hold under the stated axioms governing the approximation, with explicit epsilon-tracking in the arguments.

Significance. If the central construction is sound, the work extends classical classification theory to approximate logics in a way that preserves key structural equivalences. This could be useful for connecting discrete model theory with continuous or metric settings. The explicit epsilon-tracking and verification that standard proofs adapt without hidden circularity or inconsistency are strengths; the paper ships a coherent transfer of the classical theorems rather than ad-hoc modifications.

minor comments (3)
  1. §2: the definition of the tolerance relation on formulas should include an explicit statement of whether it is required to be symmetric and transitive or only reflexive, as this affects the subsequent type space topology.
  2. §4.2, after Definition 4.5: the statement that forking is preserved under small perturbations would benefit from a short remark on the choice of epsilon-delta constants to avoid ambiguity in the limit case.
  3. The paper would be strengthened by adding a brief comparison paragraph in the introduction to existing work on continuous logic or approximate theories (e.g., references to Ben Yaacov et al. or Henson's work on metric structures).

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their careful reading of the manuscript and for the positive assessment of the framework for approximate stability and simplicity. We are pleased that the explicit epsilon-tracking and the preservation of the classical equivalences are viewed as strengths. We will incorporate minor revisions to address any small presentational or technical points in the next version.

Circularity Check

0 steps flagged

No significant circularity; framework and proofs are self-contained

full rationale

The paper constructs an approximate first-order logic by defining formulas and types via a metric or tolerance relation, then explicitly re-derives forking, dividing, and classical equivalences (stability without order property, simplicity without TP2) while tracking epsilons in the proofs. These steps follow the standard model-theoretic arguments with the new axioms as independent inputs rather than self-definitions or fitted parameters. No load-bearing self-citations, uniqueness theorems imported from prior author work, or ansatzes smuggled via citation are used; the generalizations hold by direct verification under the stated approximation axioms, rendering the derivation chain independent of its own outputs.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Only the abstract is available, so no free parameters, axioms, or invented entities can be extracted from the full manuscript.

pith-pipeline@v0.9.0 · 5510 in / 921 out tokens · 31729 ms · 2026-05-18T17:36:29.572951+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

17 extracted references · 17 canonical work pages

  1. [1]

    J. T. Baldwin and A. H. Lachlan, On strongly minimal sets, J. Symbolic Logic36(1971), 79–96; MR0286642

    work page 1971

  2. [2]

    C. C. Chang and H. J. Keisler,Model theory, third edition, Studies in Logic and the Foundations of Mathematics, 73, North-Holland, Amsterdam, 1990; MR1059055

    work page 1990

  3. [3]

    Chernikov and S

    A. Chernikov and S. Starchenko, A note on the Erd˝ os-Hajnal property for stable graphs, Proc. Amer. Math. Soc.146 (2018), no. 2, 785–790; MR3731711

    work page 2018

  4. [4]

    Countable Ramsey. Technical Report,

    L. Coregliano and M. Malliaris, “Countable Ramsey. Technical Report,” arXiv preprint (2024); arXiv:2203.10396

    work page arXiv 2024

  5. [5]

    On the Regularity of Almost Stable Relations,

    M. Gir´ on, “On the Regularity of Almost Stable Relations,” arXiv preprint (2025); arXiv:2508.00511

    work page arXiv 2025

  6. [6]

    P. R. Halmos,Lectures on Boolean algebras, Van Nostrand Mathematical Studies, No. 1, D. Van Nostrand Co., Inc., Princeton, NJ, 1963; MR0167440

    work page 1963

  7. [7]

    Hodges,Model theory, Encyclopedia of Mathematics and its Applications, 42, Cambridge Univ

    W. Hodges,Model theory, Encyclopedia of Mathematics and its Applications, 42, Cambridge Univ. Press, Cambridge, 1993; MR1221741

    work page 1993

  8. [8]

    Hrushovski, Approximate equivalence relations, Model Theory3(2024), no

    E. Hrushovski, Approximate equivalence relations, Model Theory3(2024), no. 2, Special issue on the occasion of the 75th birthday of Boris Zilber, 317–416; MR4776304

    work page 2024

  9. [9]

    Hrushovski, Stable group theory and approximate subgroups, J

    E. Hrushovski, Stable group theory and approximate subgroups, J. Amer. Math. Soc.25(2012), no. 1, 189–243; MR2833482

    work page 2012

  10. [10]

    Kim, Forking in simple unstable theories, J

    B. Kim, Forking in simple unstable theories, J. London Math. Soc. (2)57(1998), no. 2, 257–267; MR1644264

    work page 1998

  11. [11]

    Malliaris and S

    M. Malliaris and S. Shelah, Regularity lemmas for stable graphs, Trans. Amer. Math. Soc.366(2014), no. 3, 1551–1585; MR3145742

    work page 2014

  12. [12]

    Malliaris and A

    M. Malliaris and A. Pillay, The stable regularity lemma revisited, Proc. Amer. Math. Soc.144(2016), no. 4, 1761–1765; MR3451251

    work page 2016

  13. [13]

    Martin–Pizarro and D

    A. Martin–Pizarro and D. Palacin and J. Wolf, Stability, corners, and other 2-dimensional shapes, arXiv preprint (2022); arXiv:2210.14039

    work page arXiv 2022

  14. [14]

    M. D. Morley, Categoricity in power, Trans. Amer. Math. Soc.114(1965), 514–538; MR0175782

    work page 1965

  15. [15]

    S. Shelah,Classification theory and the number of nonisomorphic models, second edition, Studies in Logic and the Foun- dations of Mathematics, 92, North-Holland, Amsterdam, 1990; MR1083551

    work page 1990

  16. [16]

    M. H. Stone, The theory of representations for Boolean algebras, Trans. Amer. Math. Soc.40(1936), no. 1, 37–111; MR1501865

    work page 1936

  17. [17]

    Tent and M

    K. Tent and M. Ziegler,A course in model theory, Lecture Notes in Logic, 40, Assoc. Symbol. Logic, La Jolla, CA, 2012 Cambridge Univ. Press, Cambridge, 2012; MR2908005 Logic Group, University of California, Berkeley, 910 Evans Hall, Berkeley, CA 94720, United States Email address:burka@berkeley.edu

    work page 2012