Regime-Boundary Sweep¶
Test or assessment — instantiates Universality Extraction
Varies scale, intensity, coupling, population, environment, or mechanism regime to locate where a macro-invariant weakens, changes form, or fails.
Every invariant is validated in some interior — a comfortable band of scale, load, and coupling where it was first seen to hold. Regime-Boundary Sweep deliberately leaves that interior, pushing the operating conditions outward until the pattern bends, changes mechanism, or breaks, and recording where. Its defining move is that it manipulates the regime the whole case sits in — turning up the scale, the intensity, the population size, the coupling strength — rather than knocking out details within a fixed setup. The output it cares about is not "does the pattern hold?" but "where does it stop holding, and what happens on the far side?" A universality claim without a mapped boundary is just an untested interior wearing a confident label.
Example¶
A controls engineer has an invariant for a fleet of grid-tied battery inverters: a droop-control relation keeps the fleet's frequency response stable. It holds beautifully in the tested interior — a few dozen units, moderate load. The Regime-Boundary Sweep drives the conditions past that interior. It scales up the number of coupled inverters and finds that past a few hundred units, communication latency turns the once-stable response oscillatory — a regime change, not a mere degradation. It increases coupling between units and locates a threshold where the stable relation flips into a resonance mode. It stresses the environment (a weak grid with high impedance) and finds the invariant simply fails.
Each breakpoint becomes an entry on a map: stable to ~200 units on a stiff grid; oscillatory beyond; invalid on weak grids. The high-stress failures are logged as first-class edge-regime counterexamples, not swept under the rug — the weak-grid failure in particular is a case that shares the structure but produces the opposite outcome, exactly the kind of adversarial evidence that keeps the class honest. The transfer advice that ships with the invariant is now conditional: safe to reuse below the mapped thresholds, redesign required above them.
How it works¶
- Pick the regime axes. Choose the dimensions — scale, intensity, coupling, population, environment, mechanism — most likely to change the pattern, and set a sweep range beyond the validated interior.
- Sweep and watch for change of kind. Move each axis and detect not just weakening but qualitative transitions: onset of oscillation, a mechanism switch, an abrupt collapse.
- Locate and log breakpoints. Pin where the invariant bends or fails, and capture the far-side behavior as an edge-regime counterexample.
- Attach evidence to each boundary. Tie every mapped limit to the observation that established it, so users can see why the boundary sits where it does.
Tuning parameters¶
- Sweep range — how far past the validated interior you push. Wider ranges find real limits but cost effort and can enter conditions no user will meet.
- Axis selection — which regime dimensions to vary. Sweeping the wrong axes gives false confidence; the axes should be the ones a transferer might actually move into.
- Resolution near breakpoints — how finely you sample around a suspected transition. Fine resolution pins the boundary precisely; coarse sampling risks straddling a regime change unseen.
- Change-of-kind sensitivity — how alert the sweep is to qualitative shifts (mechanism change) versus smooth decline. Set too blunt, it reports a gentle fade where a regime actually flipped.
- Unknown-region policy — how conservatively you label regimes you did not reach: presumed-valid, presumed-invalid, or explicitly untested.
When it helps, and when it misleads¶
Its strength is that it converts an invariant into bounded guidance: instead of "this holds," it delivers "this holds up to here, changes form there, fails beyond" — the difference between a claim that travels safely and one that leaks into conditions it was never tested in.[n1] It is the mechanism that catches regime changes an interior study can never see.
Its central failure mode is the one it exists to prevent — scale leakage, carrying a class outside its tested regime — when the sweep is too narrow or omits the axis that actually matters, leaving a boundary unmapped and later crossed unknowingly. The classic misuse is declaring an invariant "universal" from interior evidence alone, with no sweep at all, so the first out-of-regime user discovers the boundary the hard way. The guarding discipline is to sweep specifically the axes a transferer is likely to move along, and to label unreached regimes as explicitly unknown rather than silently assuming the pattern extends into them.
How it implements the components¶
transfer_limit_map— it is the map: the demonstrated-valid interior, the adaptation zones where the pattern changes form, the failure regimes, and the unknown regions, each tied to the evidence that placed it.adversarial_counterexample_set— the far side of each breakpoint yields edge-regime counterexamples (same structure, different or failed outcome) that it captures as first-class evidence rather than exceptions.
It does not knock out within-case details to test their incidental status (microdetail_perturbation_plan) — that is Microdetail Ablation Suite, its nearest test-type twin; the separation is that the sweep moves the whole case across regimes while ablation manipulates features inside a fixed regime. It also does not state the pattern (candidate_macro_invariant) or govern the membership rule (universality_class_membership_rule).
Related¶
- Instantiates: Universality Extraction — it supplies the boundary and failure-regime evidence that bounds every transfer claim.
- Consumes: Invariant Signature Induction supplies the candidate invariant whose regime limits it charts.
- Sibling mechanisms: Maximum-Variation Case Sampling · Relational Case Normalization · Invariant Signature Induction · Microdetail Ablation Suite · Red-Team Case Search · Equivalence-Class Refinement
Draft — one mechanism instantiating part of the Universality Extraction archetype; templated operating steps and generic inputs live on the archetype page.
Editorial Notes¶
Form Classification¶
Form family: Experiment, Test & Rehearsal
Rationale: Regime-Boundary Sweep operates as an active test, trial, simulation, drill, or rehearsal that generates evidence through a deliberate attempt or perturbation because it varies scale, intensity, coupling, population, environment, or mechanism regime to locate where a macro-invariant weakens, changes form, or fails.
Independent corroboration: The frozen evidence defines Regime-Boundary Sweep as 'Varies scale, intensity, coupling, population, environment, or mechanism regime to locate where a macro-invariant weakens, changes form, or fails', so its operative form is Experiment, Test & Rehearsal.
Nearest alternative: Analysis, Modeling & Optimization — Regime-Boundary Sweep includes features of an analytical, modeling, inference, comparison, or optimization procedure that derives insight or a solution, but its defining operation is an active test, trial, simulation, drill, or rehearsal that generates evidence through a deliberate attempt or perturbation.
Review outcome: Independent reviewer agreement; medium confidence.
Origin Attribution¶
Primary origin: Physics
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Universal
Rationale: Sweeping scale and coupling to locate universality regimes is rooted in statistical physics.
Related originating lineages:
- Mathematics — Asymptotic and bifurcation analysis materially formalize regime boundaries.
- Statistics & Experimental Design — Designed parameter sweeps contribute empirical detection of failure transitions.
Review resolution: Both blind reviewers agree that physics is the primary origin. Explicit reconciliation of alternate origin disagreement, domain reach disagreement adopts reviewer_a's classification because sweeping scale and coupling to locate universality regimes is rooted in statistical physics. The resulting lineage records alternates=mathematics, statistics_experimental_design, origin_mode=cross_disciplinary_synthesis, and domain_reach=universal; these describe formative provenance separately from later applicability.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
[n1] Change-point analysis is the statistical detection of a moment where a process's underlying regime shifts — its mean, variance, or generating mechanism changes rather than merely drifting. It is the quantitative counterpart of what a boundary sweep hunts for: not gradual decline but the point where the pattern becomes a different kind of thing. ↩