pith. sign in

arxiv: 1906.08668 · v1 · pith:HEPS5XDJnew · submitted 2019-06-20 · 🧮 math.PR

The Schelling model on mathbb{Z}

Maria Deijfen , Timo Hirscher This is my paper

Pith reviewed 2026-05-25 19:06 UTC · model grok-4.3

classification 🧮 math.PR
keywords Schelling modelKawasaki dynamicsmoving distributionbounded supportunbounded supportlazy agentsasymptotic behaviorinfinite lattice
0
0 comments X

The pith

The Schelling model on Z exhibits different asymptotic behavior depending on whether the moving distribution has bounded or unbounded support and on whether agents swap only on strict improvement.

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

The paper defines a Schelling model on the integers where two types of agents swap with an opposite-type agent selected by a moving distribution, provided the swap benefits both. It shows that the long-term configuration changes when the distribution allows only finite-range moves versus moves of arbitrary length, and changes again when agents are restricted to swaps that strictly improve their local situation rather than allowing non-worsening ones. This supplies a rigorous analysis of Kawasaki dynamics on an infinite lattice with local preferences. A reader would care because the rules determine whether local type preferences produce global segregation or leave mixed regions over time.

Core claim

In this version of the Schelling model on Z, agents request swaps according to a given moving distribution and perform them if beneficial for both parties. The model exhibits different asymptotic behavior depending on whether the moving distribution has bounded or unbounded support. The behavior changes if the agents are lazy in the sense that they only swap when the situation strictly improves. Generalizations to multiple types are discussed.

What carries the argument

The moving distribution (bounded versus unbounded support) together with the lazy strict-improvement swap rule, which together fix the asymptotic regime of the Kawasaki dynamics.

If this is right

  • The process reaches qualitatively different limiting configurations when moves are restricted to a finite range than when arbitrarily long moves are possible.
  • Requiring a strict improvement for each swap produces different equilibria than allowing any non-worsening swap.
  • The distinction between bounded and unbounded support persists when the model is extended to three or more agent types.

Where Pith is reading between the lines

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

  • In spatial segregation models the tail behavior of the relocation distribution may decide whether complete local clustering is reached or whether mixed interfaces persist.
  • The strict-improvement rule can be viewed as a discrete energy-decreasing dynamics whose stationary states differ from those of a non-strict version.
  • Analogous dependence on interaction range is likely to appear in other lattice particle systems with local preferences.

Load-bearing premise

The support of the moving distribution and the requirement of strict improvement for swaps are the parameters that control the qualitative long-term behavior of the process.

What would settle it

Simulate the dynamics starting from a random configuration using a uniform distribution on moves of size at most 1 and compare the eventual density of unlike-neighbor pairs to the same quantity obtained with a distribution that has positive mass on arbitrarily large jumps.

Figures

Figures reproduced from arXiv: 1906.08668 by Maria Deijfen, Timo Hirscher.

Figure 1
Figure 1. Figure 1: In the two-color the configuration is described by the separators. [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: An illustration of the event CN (t): two monochromatic sections merge. We claim that P [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: “Autonomous moving” is only possible for monochromatic sections of length at most [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Coalescence ends the timeline of two monochromatic sections. [PITH_FULL_IMAGE:figures/full_fig_p012_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: With more than two types of agents, separators do not move in simple random walks [PITH_FULL_IMAGE:figures/full_fig_p015_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Forcing separators to coalesce (by means of local modification) works with more than [PITH_FULL_IMAGE:figures/full_fig_p017_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: A configuration with c = 3 where certain sites may not converge in a finite system due to repeated singleton jumps. 17 [PITH_FULL_IMAGE:figures/full_fig_p017_7.png] view at source ↗
read the original abstract

A version of the Schelling model on $\mathbb{Z}$ is defined, where two types of agents are allocated on the sites. An agent prefers to be surrounded by other agents of its own type, and may choose to move if this is not the case. It then sends a request to an agent of opposite type chosen according to some given moving distribution and, if the move is beneficial for both agents, they swap location. We show that certain choices in the dynamics are crucial for the properties of the model. In particular, the model exhibits different asymptotic behavior depending on whether the moving distribution has bounded or unbounded support. Furthermore, the behavior changes if the agents are lazy in the sense that they only swap location if this strictly improves their situation. Generalizations to a version that includes multiple types are discussed. The work provides a rigorous analysis of so called Kawasaki dynamics on an infinite structure with local interactions.

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

1 major / 2 minor

Summary. The manuscript defines a two-type Schelling model on the integers via Kawasaki swap dynamics driven by a moving distribution. Agents request swaps with opposite-type agents; a swap occurs if it is (weakly or strictly) beneficial for both. The central claims are that the long-run behavior changes qualitatively according to whether the moving distribution has bounded versus unbounded support, and further changes when agents are required to be 'lazy' (swap only on strict improvement). Generalizations to multiple types are sketched.

Significance. The work supplies a rigorous treatment of an infinite-volume interacting particle system whose microscopic rules are shown to control macroscopic segregation. Establishing distinct asymptotic regimes for bounded/unbounded support and lazy/non-lazy rules is a substantive contribution to the mathematical study of segregation models, provided the underlying Markov process is well-defined in all cases.

major comments (1)
  1. [Definition of the dynamics and process construction] The construction of the Markov process itself is load-bearing for all claims about unbounded-support distributions. When the moving distribution has unbounded support, the total exit rate from a configuration is an infinite sum of positive terms; the manuscript must supply an explicit argument (graphical construction, domain of the generator, or non-explosion criterion) showing that the process is nevertheless well-defined on the infinite line before any long-time analysis can proceed. This point is not addressed in the abstract and appears to be the weakest link in the argument.
minor comments (2)
  1. [Model definition] The precise definition of 'beneficial' (weak versus strict improvement) should be stated once at the outset and used consistently in all statements of the main theorems.
  2. [Main results] Statements of the main theorems should include the precise mode of convergence (almost-sure, in probability, in distribution) and the topology on configurations.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful reading and for identifying the need to make the Markov process construction fully explicit, especially for moving distributions with unbounded support. We address the comment below and will revise the manuscript accordingly.

read point-by-point responses

  1. Referee: The construction of the Markov process itself is load-bearing for all claims about unbounded-support distributions. When the moving distribution has unbounded support, the total exit rate from a configuration is an infinite sum of positive terms; the manuscript must supply an explicit argument (graphical construction, domain of the generator, or non-explosion criterion) showing that the process is nevertheless well-defined on the infinite line before any long-time analysis can proceed. This point is not addressed in the abstract and appears to be the weakest link in the argument.

    Authors: We agree that an explicit construction and non-explosion argument must be supplied for the unbounded-support case, as the total rate can indeed be infinite. The original manuscript constructs the process via independent Poisson clocks attached to each ordered pair of sites (with rates given by the moving distribution) and uses the local utility function to determine acceptance, but the non-explosion step is only sketched. In the revision we will add a dedicated subsection providing the full graphical construction together with a proof that the number of jumps in any finite interval is almost surely finite; the argument relies on the fact that only finitely many sites can influence a given finite window in finite time and on standard comparison with a dominating Poisson process whose intensity is controlled by the moving distribution. revision: yes

Circularity Check

0 steps flagged

No circularity: theorems on long-run behavior of defined Markov process are independent of inputs

full rationale

The paper defines a specific Kawasaki swap dynamics on Z with given moving distribution and laziness rule, then proves theorems on its asymptotic behavior (segregation or lack thereof) depending on whether support is bounded/unbounded and whether swaps are strict-improvement only. These are statements about the constructed process, not reductions of outputs to fitted parameters, self-citations, or ansatzes by construction. No load-bearing uniqueness theorem is imported from prior self-work, and no prediction is statistically forced by a fit to the same data. The derivation chain consists of standard Markov process analysis (generator, graphical construction, coupling) applied to the explicitly stated rules; it is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Only the abstract is available. The model definition implicitly relies on standard probability axioms for the existence of the Markov process on Z; no free parameters or invented entities are visible in the abstract.

axioms (1)
  • standard math Existence and well-posedness of the continuous-time Markov chain on configurations of two types on Z with the given swap rates.
    Invoked implicitly when the abstract states that the model is defined and its asymptotic behavior can be analyzed.

pith-pipeline@v0.9.0 · 5675 in / 1281 out tokens · 23263 ms · 2026-05-25T19:06:50.240978+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

16 extracted references · 16 canonical work pages

  1. [1]

    and Lewis-Pye, A

    Barmpalias, G., Elwes, R. and Lewis-Pye, A. : Digital morphogenesis via Schelling segregation, in FOCS 2014 55th Annual IEEE Symposium on Foundations of Computer Science, 2014

    work page 2014

  2. [2]

    and Lewis-Pye, A

    Barmpalias, G., Elwes, R. and Lewis-Pye, A. : Tipping points in 1-dimensional Schelling models with switching agents, J. Stat. Phys. 158, 806-852, 2015

    work page 2015

  3. [3]

    and Lewis-Pye, A

    Barmpalias, G., Elwes, R. and Lewis-Pye, A. : Unperturbed Schelling segregation in two or three dimensions, J. Stat. Phys. 164, 1460-1487, 2016

    work page 2016

  4. [4]

    and Randall, D.: Clustering and mixing times for segrataion models on Z2, in Proceedings of the 25th Annual ACM-SIAM Symposium on Discrete Algo- rithms, SODA 2014 , 327-340, 2014

    Bhakta, P., Sarah, M. and Randall, D.: Clustering and mixing times for segrataion models on Z2, in Proceedings of the 25th Annual ACM-SIAM Symposium on Discrete Algo- rithms, SODA 2014 , 327-340, 2014

    work page 2014

  5. [5]

    and Kleinberg, R

    Brandt, C., Immorlica, N., Kamath, G. and Kleinberg, R. : An analysis of one- dimensional Schelling segregation, in Proceedings of the 44th Annual ACM Symposium on Theory of Computing , 2012

    work page 2012

  6. [6]

    and Schramm, O.: Group-invariant percolation on graphs,GAFA 9, 29-66, 1999

    Benjamini, I., Lyons, R., Peres, Y. and Schramm, O.: Group-invariant percolation on graphs,GAFA 9, 29-66, 1999

    work page 1999

  7. [7]

    Dj¨akneg˚ard, J.: Schellings segregationsmodell – en simuleringsstudie, Bachelor Thesis in Mathematical Statistics, Stockholm University, 2015

    work page 2015

  8. [8]

    Durrett, R.: Probability: Theory and Examples, 4th edition, Cambridge University Press, 2010

    work page 2010

  9. [9]

    and Sheffield, S.: Scaling limits of the Schelling model, to appear in Probab

    Holden, N. and Sheffield, S.: Scaling limits of the Schelling model, to appear in Probab. Theory Related Fields, 2018

    work page 2018

  10. [10]

    Immorlica, N., Kleinberg, R., Lucier, B. and Zadimoghaddam, M.: Exponential segregation in a two-dimensional Schelling model with tolerant individuals, in Proceedings of the 28th Annual ACM-SIAM Symposium on Discrete Algorithms, SODA 17 , 984-993, 2017

    work page 2017

  11. [11]

    Kallenberg, O.: Foundations of Modern Probability (2nd edition) , Springer, 2002

    work page 2002

  12. [12]

    and Weiss, H.: The dyanmics of Schelling-type segregation models and a nonlinear graph Laplacian variational problem, Adv

    Pollicott, M. and Weiss, H.: The dyanmics of Schelling-type segregation models and a nonlinear graph Laplacian variational problem, Adv. Appl. Math. 27, 17-40, 2001

    work page 2001

  13. [13]

    Schelling, T.: Models of segregation, Am. Econ. Rev. 59, 488-493, 1969

    work page 1969

  14. [14]

    Schelling, T.: Dynamic models of segregation, J. Math. Sociol. 1, 143-186, 1971

    work page 1971

  15. [15]

    Young, H.: Individual strategy and social structure: An evolutionary theory of institutions, Princeton University Press, 1998

    work page 1998

  16. [16]

    Zhang, J.: A dynamic model of residential segregation, J. Math. Sociol. 28, 147-170, 2004. 18

    work page 2004