Borel–Padé Resummation¶

Method — instantiates Resummation and Nonperturbative Extrapolation

Continues a Borel-transformed series by rational (Padé) approximation from finitely many coefficients, reading its Borel-plane poles to reconstruct a divergent expansion.

Version: v1 · 2026-08-24 · History
Mechanism #: 877
Type: Method
Form family: Analysis, Modeling & Optimization
Solution family: Substitution & Fallback
Problem family: Correctness, Conformance & Formal Validity Failure
Problem subfamily: Computational Decidability & Bounded Approximation
Origin domain: Mathematics
Also from: Physics
Instantiates: Resummation and Nonperturbative Extrapolation

You have a handful of perturbation coefficients that grow factorially, so summing them directly is hopeless — but the Borel transform, formed by dividing each coefficient by an order-dependent factorial, tames that growth into a series that might actually converge. The catch is that you only know finitely many Borel coefficients, so you cannot integrate the Borel function until you continue it past where the terms run out. Borel–Padé Resummation does that continuation with a rational (Padé) approximant built on the Borel series, then reads the approximant's poles as candidate singularities in the Borel plane. Its defining move lives entirely in the Borel plane: it never touches the physical variable with a rational fit and never assumes the Borel function is known in closed form. It transforms first, records the transformed object, then rationally continues it.

Example¶

A precision calculation of the electron's anomalous magnetic moment in quantum electrodynamics produces perturbative coefficients whose magnitudes climb roughly like a factorial once you pass the first few orders — the series is asymptotic, not convergent. Suppose eight coefficients are in hand. Direct partial sums improve, then visibly degrade around the fifth term. The analyst forms the Borel transform (coefficient n divided by n!), which now looks like a well-behaved short series, and replaces it with a near-diagonal Padé approximant so it can be evaluated past the eighth term. The approximant develops a pole on the negative Borel axis at roughly a fixed distance — the same location across several order pairs — which is exactly where a singularity is expected on physical grounds. Integrating the continued Borel function back yields a finite reconstructed value with a small, honest spread across the order grid, where naive summation gave a number that wandered with every added term.

How it works¶

Diagnose the growth. Estimate the coefficient ratios and confirm genuine factorial (or near-factorial) growth; that is what licenses the Borel step rather than, say, a raw sequence transform.
Form and record the Borel series. Divide out the factorial and treat the result as a new expansion object with its own normalization and weight parameter.
Rationally continue in the Borel plane. Fit near-diagonal [L/M] approximants to the Borel series so it can be evaluated where the finite coefficients stop.
Sweep and integrate. Build a small grid of order pairs, keep only Borel-plane poles stable across it, and integrate the surviving continuation back to a value.

Tuning parameters¶

Borel–Padé order pair [L/M] — numerator and denominator degrees of the rational fit to the Borel series. Higher orders track more structure but chase noise in the last coefficients.
Near-diagonal vs. off-diagonal — how far from L≈M you stray. Diagonal shapes are usually most stable; off-diagonal shapes probe how much the result leans on shape choice.
Borel weight — the power of the factorial divided out (ordinary Borel vs. a generalized weight). A heavier weight tames stronger growth but can over-smooth.
Order-grid span — how many [L/M] combinations enter the sweep. A wider span separates stable from order-dependent Borel-plane features but costs interpretation time.

When it helps, and when it misleads¶

Its strength is turning a genuinely divergent asymptotic series into a controlled value using only the coefficients you have, while making the Borel-plane singularity structure visible rather than assumed — a divergence that Dyson's argument says is intrinsic to the theory, not a numerical accident.^[1] Because the continuation is a cheap rational fit, you can afford a whole grid of them and let stability do the arguing.

Its central failure is the spurious Borel-plane pole: a Padé approximant will happily invent a pole that looks like a physical singularity but jumps when one coefficient is removed, and a value read off such a pole is an artifact. False convergence is the subtler trap — the first several order pairs can agree before the true large-order regime begins, so a reassuring stability plot proves less than it seems. The guarding discipline is to trust only Borel-plane features that survive an informal leave-one-out check across the order grid, and to treat any pole that moves under a dropped coefficient as noise until something independent supports it.

How it implements the components¶

large_order_or_singularity_diagnostic — the ratio/growth read that both justifies the Borel transform and marks the candidate Borel-plane singularity the continuation must respect.
local_expansion_record — the Borel-transformed series (coefficients ÷ factorial, with its weight and normalization) recorded as the evidence object the rational step consumes.
continuation_ensemble — the grid of [L/M] order pairs whose agreement separates real Borel-plane structure from shape-dependent artifacts.

It does not perform the inverse-Borel integral as a governed choice or own its contour ambiguity — the resummation_transform_choice, ambiguity_and_uncertainty_budget, and nonperturbative_completion_boundary are Borel Resummation's, its nearest twin — and the auditable artifact_rejection_log of excluded poles belongs to Padé Approximant.

Instantiates: Resummation and Nonperturbative Extrapolation — supplies a usable continuation of the Borel transform when only finitely many coefficients exist.
Consumes: Padé Approximant — the rational-continuation machinery, here applied in the Borel plane rather than the physical variable.
Sibling mechanisms: Borel Resummation · Conformal Borel Mapping · Exact-or-Numerical Benchmark · High-Temperature Series Resummation · Matched Asymptotic Expansion · Padé Approximant · Renormalization-Group Improvement · Sequence Acceleration Transform · Strong-Coupling Extrapolation Check

Editorial Notes¶

Form Classification¶

Form family: Analysis, Modeling & Optimization

Rationale: Continues a Borel-transformed series by rational (Padé) approximation from finitely many coefficients, reading its Borel-plane poles to reconstruct a divergent expansion, making its operative form a computation or analytic transformation that produces an inference, comparison, or optimized result.

Independent corroboration: The frozen evidence defines Borel–Padé Resummation as 'Continues a Borel-transformed series by rational (Padé) approximation from finitely many coefficients, reading its Borel-plane poles to reconstruct a divergent expansion', so its operative form is Analysis, Modeling & Optimization.

Review outcome: Independent reviewer agreement; high confidence.

Origin Attribution¶

Primary origin: Mathematics

Origin pattern: Single lineage

Present-day reach: Specialized

Rationale: Mathematical analysis combines the Borel transform with Padé rational approximation to analytically continue a factorially divergent series in the Borel plane before inversion.

Related originating lineages:

Physics — Physics contributes the wave, boundary-condition, scaling, perturbation, or measurement formalism used here.