Equilibrium decisions under partial observation: dynamical error bounds and information requirements

arXiv:2607.14815 · cs.MA · Submitted 2026-07-16 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Computer Science. The day in computer science: software, systems and theory on arXiv.

Tom: We're looking at the paper: "Equilibrium decisions under partial observation".

Paper summary: Tom: So to recap, this paper, "Equilibrium decisions under partial observation: dynamical error bounds and information requirements," is formally about urban mobility where you can only observe some indicators, and it builds a multi-layer dynamic attractor.

Jane: They present a dynamic multi-layer equilibrium attractor for disrupted urban mobility, which has a fast performance layer relaxing to an indicator-dependent target, and a slow strategic layer providing the joint fixed point of traffic assignment, modal split, and capacity learning >

Lu: The core thesis is about multi-scale equilibrium reducibility under variable indicator dimensionality. They're treating that data mismatch between what's in the model and what you actually have formally >

Meng: So, what are they claiming? They claim they can rigorously characterize when a lower-dimensional indicator projection is faithful to this structure, which means checking if the reduced system still behaves like the original one >

Tom: They characterize this faithfulness through four results: exact and approximate projectability with error bounds covering non-normal transient amplification, preservation of the coupled two-layer fixed point up to a contraction boundary with a displacement bound under ap, and statistical decision power for antifragility >

Jane: That sounds like they are giving us the mathematical tools to say when we can trust simplified models based on incomplete data. It’s about defining what's reliable when observation is partial >

Conclusion: Tom: Thinking about the title, "Equilibrium decisions under partial observation," it really highlights that the problem isn't just about having bad data; it’s about how your decisions—like setting traffic signals or rerouting buses—are affected when you only see a piece of the picture >

Jane: And this paper shows that we can go beyond just hoping a model works. They provide concrete conditions, like projectability and commutation, that tell us precisely when our simplified model will still capture the true underlying system dynamics >

Lu: It’s about moving from just making an approximation to having a formal proof of how good that approximation is under real-world constraints like incomplete data availability >

Meng: For those of us shipping models, this means we can be more confident in the models we deploy, knowing exactly where the mathematical guarantees break down when we start compressing the indicator space >

Tom: And for policy makers or city planners, it means understanding that if you only measure throughput but ignore equity data, your equilibrium estimates might be fundamentally flawed because of that missing link >

Jane: It’s about formalizing the trade-off: less data means simpler models, but we need to know exactly what error bound you are accepting and what that error actually is when monitoring things with fewer indicators >

Ali Ghoroghi, Yacine Rezgui, Afrouz Ghaemi, Cristina De Nardi, Andrei Hodorog

School of Engineering, Cardiff University

cs.MA

Submitted: 2026-07-16

Updated: 2026-11-17

Comments: 23 pages, 6 figures, 6 tables. v4: substantially revised and corrected; title changed, scope narrowed, phase-initialisation statement corrected

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 92/100

The gist: Equilibrium analysis of urban mobility is formulated in a high-dimensional indicator space, whilst data availability varies sharply across cities and disruptions.

Key concepts

Multi-scale Equilibrium Reducibility
This is the theory that allows complex urban mobility equilibrium analysis to be simplified. It involves separating the system into a fast performance layer and a slow strategic layer, which helps manage high-dimensional indicator spaces by finding lower-dimensional representations that still capture essential dynamics.
Faithful Reduction
This is the formal contribution defining what it means for a data reduction method to be successful. It requires preserving not just the trajectory but also the stability class of the system's performance layer and ensuring that statistical tests remain powerful under the reduced model.
Projectability and Hurwitz Property
These are mathematical conditions used to guarantee that simplified models accurately reflect the original system's stability. Projectability ensures that a vector field can be projected onto a lower-dimensional space while maintaining certain dynamical properties, which is crucial for proving the reduced model's reliability.
Restoration Time Bias
This result shows that monitoring based on a reduced set of indicators will always optimistically underestimate how long it takes for the entire urban system to return to its true equilibrium state. Reduced monitoring can only understate recovery duration.

Terminology

Summary

Equilibrium analysis of urban mobility is formulated in a high-dimensional indicator space, whilst data availability varies sharply across cities and disruptions. p. 0

Where the paper says it

Multi-Scale Equilibrium under Variable Indicator Dimensionality: Faithful Reduction of Dynamic Attractors in Urban Mobility Systems Ali Ghoroghi∗1 , Yacine Rezgui1 , Afrouz Ghaemi1 , Cristina De Nardi1 , and Andrei Hodorog1 1School of Engineering, Cardiff University, Cardiff, United Kingdom Abstract Equilibrium analysis of urban mobility is formulated in a high-dimensional indicator space, whilst data availability varies sharply across cities and disruptions. This paper treats that mismatch formally.

The formal contribution is a theory of multi-scale equilibrium reducibility under variable indicator dimensionality. p. 1
Where the paper says it

Having established that attractor, the paper asks when it remains meaningful if the full indicator vector is replaced by a lower-dimensional support, as occurs when different cities maintain different data infrastructures or when some measurements are unavailable during a disruption. The formal contribution is a theory of multi-scale equilibrium reducibility under variable indicator dimensionality.

The Equilibrium Framework

The paper presents a dynamic multi-layer equilibrium attractor for disrupted urban mobility, in which a fast performance layer relaxes towards an indicator-dependent target, a slow strategic layer supplies a joint traffic, modal and learning fixed point. This framework involves a performance-layer state variable evolving according to a relax-to-target ordinary differential equation. p. 3

Where the paper says it

The strategic layer produces the slow-timescale baselines that the performance layer tracks; the performance layer in turn supplies the post-event trajectories from which the strategic layer learns. 1 Performance-layer dynamics The performance state evolves according to the relax-to-target ordinary differential equation E˙(t) = −κ E(t) − E¯(x(t)) , κ > 0, (2) with discrete-time update Et+∆t = Et −κ ∆t(Et −E¯(xt)), in which κ ∆t ∈ (0, 1] preserves monotone convergence, values in (1, 2) converge with overshoot and values at or above two are unstable.

The strategic layer tracks a hierarchical fixed point combining user-equilibrium traffic assignment, modal split and capacity learning. p. 3
Where the paper says it

The fast performance layer is specified by a relaxation law on E(t); the slow strategic layer is specified by a hierarchical fixed point combining user-equilibrium traffic assignment, modal split and a capacity-learning update. The strategic layer produces the slow-timescale baselines that the performance layer tracks; the performance layer in turn supplies the post-event trajectories from which the strategic layer learns.

Antifragility is determined by a candidate decision rule comparing a post-event performance ratio with a candidate baseline-dependent threshold. p. 1
Where the paper says it

In that framework, a performance-layer state variable relaxes towards a target function defined over a vector of normalised indicators, while a slower strategic layer supplies a joint traffic, modal and learning fixed point. Antifragility is determined by a candidate decision rule comparing a post-event performance ratio with a candidate baseline-dependent threshold.

Reducibility and Faithfulness

The formal contribution is defined as faithfulness relative to the attractor rather than relative to trajectory reconstruction alone. A faithful reduction requires preserving the stability class of the performance layer, the joint two-layer fixed point and the statistical power of the candidate antifragility test up to an explicitly quantified tolerance. p. 1

Where the paper says it

Given a full indicator state x ∈ R n and a reduction Π : R n → R m, with m < n, the paper defines faithfulness relative to the attractor rather than relative to trajectory reconstruction alone. A faithful reduction is required to preserve the stability class of the performance layer, the joint two-layer fixed point and the statistical power of the candidate antifragility test up to an explicitly quantified tolerance.

The novelty lies in joining three preservation criteria that have usually been studied apart, and in attaching to them an explicit measurement model. p. 1
Where the paper says it

This definition is deliberately stronger than common projection requirements because a reduced equilibrium analysis can be numerically convenient whilst being mathematically unfaithful. The novelty lies in joining three preservation criteria that have usually been studied apart, and in attaching to them an explicit measurement model.

Preservation of Stability and Fixed Points

The paper gives a necessary and sufficient condition for preservation of local asymptotic stability under linear indicator projection, formulated through projectability of the vector field and the Hurwitz property of the reduced Jacobian. It also proves a coupled fixed-point preservation theorem for a two-layer performance-strategic map, identifying the sufficient contraction boundary at which the uniqueness certificate is lost. p. 2

Where the paper says it

Secondly, it gives a necessary and sufficient condition for preservation of local asymptotic stability under linear indicator projection, formulated through projectability of the vector field and the Hurwitz property of the reduced Jacobian, together with an approximate-projectability theorem that bounds the projected-trajectory error when projectability holds only up to a measurable defect, together with a transient-amplification extension that covers nonnormal reduced dynamics through a certified, computable amplification constant. Thirdly, it proves a coupled fixed-point preservation theorem for a two-layer performance-strategic map, identifies the sufficient contraction boundary at which the uniqueness certificate is lost, and bounds the fixed-point displacement when the reduced maps commute with the projection only approximately.

The joint fixed point preservation relies on the commutation condition where the reduced map commutes with both projections. p. 6
Where the paper says it

The argument establishes joint existence only; uniqueness and asymptotic convergence of the fully coupled iteration require the contraction condition stated above, which Theorem 4 shows must additionally be preserved under any indicator reduction if the reduced system is to inherit the joint fixed point. The contraction condition on the composed capacity map reappears in Section 4 as the quantity whose preservation under reduction is at issue: Theorem 4 identifies the boundary at which a reduced coupling matrix ceases to be contractive and joint uniqueness can be lost.

Decision-Rule Degradation and Information Requirements

The paper introduces a measurement model in which the equilibrium state variables are estimated from observable urban key performance indicators, deriving the retained Fisher information of any indicator support in closed form under Gaussian measurement noise. It proves that detection power is monotone in nested indicator supports, and shows that greedy indicator selection is exactly optimal under the independent-channel measurement model. p. 2

Where the paper says it

Fourthly, it introduces a measurement model in which the equilibrium state variables are estimated from observable urban key performance indicators, derives the retained Fisher information of any indicator support in closed form under Gaussian measurement noise, and proves that detection power is monotone in nested indicator supports. It further shows that greedy indicator selection is exactly optimal under the independent-channel measurement model.

The minimum admissible support can be approached by greedy addition of the channel with the largest innovation gain in Eq. p. 2
Where the paper says it

Fourthly, it introduces a measurement model in which the equilibrium state variables are estimated from observable urban key performance indicators, derives the retained Fisher information of any indicator support in closed form under Gaussian measurement noise, and proves that detection power is monotone in nested indicator supports. It further shows that greedy indicator selection is exactly optimal under the independent-channel measurement model.

(34).

Restoration Time Bias

The paper proves that restoration time to equilibrium estimated from a reduced indicator set never exceeds, and generically understates, the restoration time of the full system, so that reduced monitoring is optimistically biased. This one-sided bias is exact as an asymptotic ordering, whereby reduced monitoring can only understate recovery duration. p. 0

Where the paper says it

It presents a dynamic multi-layer equilibrium attractor for disrupted urban mobility, in which a fast performance layer relaxes towards an indicator-dependent target, a slow strategic layer supplies a joint traffic, modal and learning fixed point, and antifragility is classified through a statistical decision rule on the post-to-baseline performance ratio. It then characterises when a lower-dimensional indicator projection is faithful to this structure, through four results: exact and approximate projectability with error bounds covering non-normal transient amplification; preservation of the coupled two-layer fixed point up to a contraction boundary, with a displacement bound under approximate commutation; the retained Fisher information and decision power of any indicator support, with greedy selection exactly optimal for independent channels; and a one-sided restoration-time bias, made exact as an asymptotic ordering, whereby reduced monitoring can only understate recovery duration.

Summary of Expected Behavior

The protocol supports the formal claims in a limited and explicit sense. It is designed to show that the theorems distinguish four failure modes that would otherwise be conflated: absence of an autonomous projected flow beyond the tolerated defect, loss of uniqueness at the coupled fixed-point boundary, loss of statistical decision power despite stable reduced dynamics. p. 8

Where the paper says it

It does not by itself show that the reduced score corresponds to a stable equilibrium projection or that a coupled fixed point remains unique. Moreover, the loss of an indicator has two roles in the present setting: it may change the target value towards which the performance layer relaxes, and it may reduce the statistical power of the decision rule used to classify post-disruption behaviour.

The required indicators are not those that make the model detailed, but those required to preserve the specific equilibrium property being asserted. p. 1
Where the paper says it

A faithful reduction is required to preserve the stability class of the performance layer, the joint two-layer fixed point and the statistical power of the candidate antifragility test up to an explicitly quantified tolerance. This definition is deliberately stronger than common projection requirements because a reduced equilibrium analysis can be numerically convenient whilst being mathematically unfaithful.

Conclusion

The first contribution is a definition of faithful reduction for a dynamic attractor, a theorem giving necessary and sufficient conditions for preservation of local asymptotic stability under projection, and a practical-faithfulness extension that bounds the projected-trajectory error by the ratio of the projectability defect to the reduced contraction rate. The second contribution is a joint fixed-point preservation theorem for coupled performance and strategic layers, including the contraction boundary at which uniqueness may be lost and the rate at which solver effort diverges as the boundary is approached. p. 2

Where the paper says it

Secondly, it gives a necessary and sufficient condition for preservation of local asymptotic stability under linear indicator projection, formulated through projectability of the vector field and the Hurwitz property of the reduced Jacobian, together with an approximate-projectability theorem that bounds the projected-trajectory error when projectability holds only up to a measurable defect, together with a transient-amplification extension that covers nonnormal reduced dynamics through a certified, computable amplification constant. Thirdly, it proves a coupled fixed-point preservation theorem for a two-layer performance-strategic map, identifies the sufficient contraction boundary at which the uniqueness certificate is lost, and bounds the fixed-point displacement when the reduced maps commute with the projection only approximately.

The third contribution is a decision-rule degradation result expressed through a measurement model that links the equilibrium state variables to observable urban indicators, yielding the retained Fisher information in closed form, a monotone innovation-gain calculus for channel selection, and an explicit admissibility condition for the minimum indicator support. p. 2
Where the paper says it

Fourthly, it introduces a measurement model in which the equilibrium state variables are estimated from observable urban key performance indicators, derives the retained Fisher information of any indicator support in closed form under Gaussian measurement noise, and proves that detection power is monotone in nested indicator supports. It further shows that greedy indicator selection is exactly optimal under the independent-channel measurement model.

The fourth contribution is a one-sided restoration-time bias result showing that reduced monitoring can only understate, never overstate, the time the full system requires to restore equilibrium. p. 0
Where the paper says it

It then characterises when a lower-dimensional indicator projection is faithful to this structure, through four results: exact and approximate projectability with error bounds covering non-normal transient amplification; preservation of the coupled two-layer fixed point up to a contraction boundary, with a displacement bound under approximate commutation; the retained Fisher information and decision power of any indicator support, with greedy selection exactly optimal for independent channels; and a one-sided restoration-time bias, made exact as an asymptotic ordering, whereby reduced monitoring can only understate recovery duration. A simulation study on three stylised pilot-city configurations verifies each result; two observable channels suffice for the classification target where the catalogue permits.

The research questions are answered by projectability, commutation of fixed-point maps with the performance and strategic projections, reduction in the non-centrality parameter of the candidate antifragility test computable from the measurement model through Eq. p. 0
Where the paper says it

It presents a dynamic multi-layer equilibrium attractor for disrupted urban mobility, in which a fast performance layer relaxes towards an indicator-dependent target, a slow strategic layer supplies a joint traffic, modal and learning fixed point, and antifragility is classified through a statistical decision rule on the post-to-baseline performance ratio. It then characterises when a lower-dimensional indicator projection is faithful to this structure, through four results: exact and approximate projectability with error bounds covering non-normal transient amplification; preservation of the coupled two-layer fixed point up to a contraction boundary, with a displacement bound under approximate commutation; the retained Fisher information and decision power of any indicator support, with greedy selection exactly optimal for independent channels; and a one-sided restoration-time bias, made exact as an asymptotic ordering, whereby reduced monitoring can only understate recovery duration.

(33) and Eq. (34), and spectral inclusion of Eq. p. 1
Where the paper says it

The conceptual difficulty is that indicator reduction is not merely a matter of approximation error. In a standard projection problem one may seek a reduced state whose trajectory is close to the original trajectory under a chosen norm; in a standard network aggregation problem one may seek preservation of selected spectral or community properties; and in a standard composite-index problem one may seek sensitivity bounds for a scalar score.

(16) and its one-sided consequence for restoration-time reporting. p. 0
Where the paper says it

It presents a dynamic multi-layer equilibrium attractor for disrupted urban mobility, in which a fast performance layer relaxes towards an indicator-dependent target, a slow strategic layer supplies a joint traffic, modal and learning fixed point, and antifragility is classified through a statistical decision rule on the post-to-baseline performance ratio. It then characterises when a lower-dimensional indicator projection is faithful to this structure, through four results: exact and approximate projectability with error bounds covering non-normal transient amplification; preservation of the coupled two-layer fixed point up to a contraction boundary, with a displacement bound under approximate commutation; the retained Fisher information and decision power of any indicator support, with greedy selection exactly optimal for independent channels; and a one-sided restoration-time bias, made exact as an asymptotic ordering, whereby reduced monitoring can only understate recovery duration.

The paper establishes the formal conditions under which partial indicator coverage is admissible and exhibits them numerically through Experiments E1 to E5 on the stylised testbed. p. 2
Where the paper says it

The present paper therefore studies a coupled reducibility problem rather than an isolated approximation problem, and it grounds that problem in the indicator infrastructure that cities actually maintain. Of particular interest is the boundary below which partial indicator coverage ceases to support the equilibrium claim.

Data Accessibility

This is a theoretical paper. The simulation protocol of Section 5 is fully specified in the text and Table 3. p. 2

Where the paper says it

These contributions are theoretical. The numerical section specifies the simulation protocol through which they are illustrated on three stylised pilot-city configurations.

The complete analysis code reproducing every figure, table and reported value of Section 5, together with its numerical outputs and the exact loading matrices CS and noise covariances ΣS used in Experiments E1 to E5 is provided as electronic supplementary material with this submission and will additionally be archived in a public repository with a persistent identifier on acceptance.

Authors' Contributions

A.G.: conceptualisation, methodology, formal analysis, investigation, writing (original draft), writing (review and editing). p. 3

Where the paper says it

Section 3 reviews reduction theory in dynamical systems, layer aggregation in multi-layer networks, sensitivity analysis for composite indicators and the design-of-experiments view of indicator selection. Section 4 develops the formal reducibility framework and proves the principal results, including the measurement model that connects the state variables to observable urban indicators.

Y.R.: supervision, funding acquisition, writing (review and editing). p. 7
Where the paper says it

3 Literature review 3. 1 Dimension reduction in dynamical systems Dimension reduction in dynamical systems is concerned with the conditions under which the evolution of a high-dimensional state can be represented on a lower-dimensional object without losing the properties of interest.

A.Gh.: investigation, validation, writing (review and editing). p. 7
Where the paper says it

3 Literature review 3. 1 Dimension reduction in dynamical systems Dimension reduction in dynamical systems is concerned with the conditions under which the evolution of a high-dimensional state can be represented on a lower-dimensional object without losing the properties of interest.

C.D.N.: writing (review and editing). p. 7
Where the paper says it

3 Literature review 3. 1 Dimension reduction in dynamical systems Dimension reduction in dynamical systems is concerned with the conditions under which the evolution of a high-dimensional state can be represented on a lower-dimensional object without losing the properties of interest.

A.H.: software, writing (review and editing). p. 7
Where the paper says it

3 Literature review 3. 1 Dimension reduction in dynamical systems Dimension reduction in dynamical systems is concerned with the conditions under which the evolution of a high-dimensional state can be represented on a lower-dimensional object without losing the properties of interest.

The authors used generative artificial intelligence tools (Claude, Anthropic) to assist with language editing, with the preparation of the LaTeX source, with the drafting of the reproducibility code implementing the simulation protocol. p. 4
Where the paper says it

Assumption 1 is the bridge between the attractor of this section and the reducibility theory of Section 4: the projection results are stated for the indicator flow f, and the scalar performance layer inherits its reduced behaviour through the induced observable. The assumption is a modelling commitment rather than a theorem; it holds exactly for the linearised recovery dynamics used in the simulation protocol of Section 5 and approximately whenever indicator recovery is smooth on the post-stabilisation window.

Competing Interests

The authors declare no competing interests.

Funding

This work was supported by the European Union’s Horizon Europe research and innovation programme under the AntifragiCity project, Grant Agreement No. 101203052.

Acknowledgements

This manuscript builds on some internal project material from AntifragiCity project. The authors used generative artificial intelligence tools (Claude, Anthropic) to assist with language editing, with the preparation of the LaTeX source, with the drafting of candidate statements and proof sketches for the extension theorems. p. 3

Where the paper says it

The twelve quantities are computed constructs, each assembled from urban indicators that municipalities and transport operators record; the assembly rules supply the measurement loadings used in Section 4. 4, and the quantities are treated as candidate constructs unless an empirical source is attached to a specific deployment.

The authors reviewed and verified the content, including all mathematical statements, proofs, and numerical results. p. 1
Where the paper says it

indicators is observable. Dynamical systems theory contains mature results on local reduction near equilibria, including centre manifold theory, inertial manifolds, singular perturbation and Galerkin projection [Carr, 1981, Guckenheimer and Holmes, 1983, Khalil, 2002, Fenichel, 1979, Temam, 1988, Holmes et al.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

1

Improved AI can perform faithful equilibrium analysis by explicitly defining faithfulness as preserving three concurrent requirements: asymptotic stability of the performance layer, preservation of the coupled two-layer fixed point, and retention of statistical power for an antifragility decision rule.

2

The improved system can determine when a lower-dimensional indicator projection is faithful to the multi-scale structure through exact or approximate projectability with error bounds covering nonnormal transient amplification.

3

The system can classify indicator reduction as practically faithful by quantifying the projectability defect using Theorem 2, which states that a reduction is practically faithful when the logarithmic norm of the reduced Jacobian satisfies a certain condition related to the tolerance delta tol.

4

The improved AI can assess if a reduced system preserves the joint two-layer fixed point by checking if it satisfies commutation of maps and whether its reduced coupling matrix has a spectral radius less than one, as stated in Theorem 4.

5

The system can calculate how the statistical power of an antifragility decision rule degrades as the retained indicator support shrinks by deriving the retained Fisher information in closed form and using a monotone innovation-gain calculus for channel selection.

6

The AI can provide a principled basis for deciding which indicators must be maintained by identifying four failure modes—absence of autonomous flow, loss of uniqueness, loss of statistical decision power, and one-sided restoration time bias—and defining the minimum data infrastructure required to preserve the mathematical object being claimed.

Abstract

Reduced monitoring can preserve stable dynamics while changing an equilibrium decision. Motivated by heterogeneous urban-mobility indicator coverage, we study fixed linear observations of a nonlinear recovery flow coupled to a strategic fixed-point map, and ask when a decision based on the reduced model can be trusted. The main result is a joint certificate: bounds on trajectory tracking, equilibrium displacement and unresolved output dependence are assembled into one decision-margin error which, under a stated Gaussian observation model, yields a conservative one-sided test and an explicit information requirement for prescribed worst-case power. In the scalar experiment this requirement is sharp: when the effect does not exceed twice the error budget, no valid test has worst-case power above its size, however many observations are taken. A second result connects manifold-adapted observations to defect bounds through a locally invertible observation chart, while distinguishing observation initialisation from asymptotic phase initialisation. Supporting results, resting on established lumping, contraction and fixed-point theory, cover exact quotients, an observation-fibre obstruction and the dependence of recovery-time comparisons on the tolerance convention. Seven reproducible synthetic experiments examine each component. The results give conditional criteria for certifying partial monitoring and separate mathematical guarantees from empirical validation.

Sources

Related papers