pith. sign in

arxiv: 2004.05221 · v5 · submitted 2020-04-10 · 🧮 math.NT

On the distribution of addition chains

Theophilus Agama This is my paper

Pith reviewed 2026-05-24 15:45 UTC · model grok-4.3

classification 🧮 math.NT
keywords addition chainsdeterminersregulatorsgapsharmonic sumarithmetic sumbalancing problemexponentiation
0
0 comments X

The pith

Decomposing each generator in an addition chain into a determiner and regulator produces explicit identities for the chain sums and reciprocal sum.

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

The paper decomposes every generator in an addition chain for a fixed target into a determiner plus a regulator, called the gap. This decomposition supplies closed identities that express the sum of all determiners and the sum of all chain elements directly in terms of the regulator sequence. A parallel identity writes the sum of reciprocals of the chain elements using the same regulators. Lower bounds on the sums follow at once from the positivity of the regulators. The identities furnish a uniform language for comparing any two chains that share the same target and length, and the paper closes by posing the problem of balancing the arithmetic sum against the harmonic sum.

Core claim

By writing each generator as the sum of a determiner and a regulator, the paper derives closed-form expressions for the total sum of determiners, the total sum of the chain, and the harmonic sum of the chain, all parametrized by the sequence of regulators. These expressions immediately imply lower bounds once the regulators are known to be positive integers. The same parametrization permits a direct comparison of any two chains with identical length and target value.

What carries the argument

The decomposition of each generator into a determiner and a regulator (gap), which parametrizes the aggregate statistics of the chain.

If this is right

  • Explicit identities for the sum of determiners and the sum of chain elements in terms of the regulator sequence.
  • Lower bounds on these sums obtained from the positivity of the regulators.
  • An identity expressing the reciprocal sum of the chain via the same gap sequence.
  • A uniform method for comparing addition chains of fixed target and length.
  • Formulation of a balancing problem that minimizes the difference between the arithmetic sum and the harmonic sum, together with a structural decomposition of the objective.

Where Pith is reading between the lines

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

  • The regulator parametrization could be used to enumerate or optimize chains by searching over admissible gap sequences rather than over the full chain.
  • The identities may extend immediately to higher-power sums or other symmetric functions of the chain elements.
  • The balancing problem admits a natural dynamic-programming formulation once the regulator sequence is treated as the decision variable.

Load-bearing premise

The decomposition of each generator into a determiner and a regulator is valid for every addition chain and directly yields the stated identities without additional unstated conditions on the chain structure.

What would settle it

An addition chain whose element sum or reciprocal sum cannot be expressed exactly in terms of its regulator sequence as claimed.

read the original abstract

Addition chains are a classical construction for fast exponentiation and related computation problems. In this paper, we study a chain for a fixed integer $n$ by decomposing each generator into a \emph{determiner} and a \emph{regulator} (gap). This viewpoint leads to explicit identities for two aggregate statistics of the chain: the sum of the determiners and the sum of the chain elements. We then derive the corresponding lower bounds by using the positivity of the regulators. In parallel, we establish an identity for the reciprocal sum of the chain, showing how the harmonic profile of the chain can also be written in terms of the same gap sequence. These identities provide a unified way to compare addition chains of the same target and length. The paper concludes with a balancing problem that asks for the chain(s) that minimize the difference between the arithmetic sum and the harmonic sum, together with a structural decomposition of that optimization objective.

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

2 major / 0 minor

Summary. The manuscript studies addition chains for a fixed integer n by introducing a determiner-regulator (gap) decomposition of each generator. This leads to explicit identities expressing the sum of determiners, the sum of chain elements, and the reciprocal sum of the chain in terms of the gap sequence. Lower bounds follow from regulator positivity. The paper ends with a balancing problem that seeks chains minimizing the difference between the arithmetic sum and harmonic sum, together with a structural decomposition of that objective.

Significance. If the decomposition is unambiguously defined and the identities hold for arbitrary addition chains, the framework supplies a unified algebraic description of aggregate statistics that could facilitate direct comparison of chains sharing the same target and length. The explicit identities (rather than purely existential bounds) and the formulation of the balancing problem are constructive strengths.

major comments (2)
  1. [Abstract / decomposition definition] The central construction decomposes each generator into a determiner and regulator without stating a canonical selection rule. Addition chains routinely admit multiple distinct pairs (j,k) with a_i = a_j + a_k. Absent an explicit tie-breaking convention or a proof that the resulting sums are independent of the choice, the claimed identities for the sum of determiners, sum of elements, and reciprocal sum cannot be asserted to hold unconditionally for every chain (see abstract and the paragraph introducing the decomposition).
  2. [Identities and lower bounds] The lower bounds obtained from regulator positivity and the subsequent balancing problem both presuppose that the gap sequence is intrinsic to the chain. If different decompositions of the same chain produce different gap sequences, the identities and the optimization objective become decomposition-dependent rather than chain-dependent, weakening the comparison claim for chains of equal length and target.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments on the decomposition. We address the major comments point by point below.

read point-by-point responses

  1. Referee: [Abstract / decomposition definition] The central construction decomposes each generator into a determiner and regulator without stating a canonical selection rule. Addition chains routinely admit multiple distinct pairs (j,k) with a_i = a_j + a_k. Absent an explicit tie-breaking convention or a proof that the resulting sums are independent of the choice, the claimed identities for the sum of determiners, sum of elements, and reciprocal sum cannot be asserted to hold unconditionally for every chain (see abstract and the paragraph introducing the decomposition).

    Authors: We agree that the manuscript does not provide an explicit tie-breaking rule for non-unique representations a_i = a_j + a_k. In the revision we will introduce a canonical convention (e.g., lexicographically smallest (j,k)) and verify that the three aggregate identities remain valid under this rule, thereby making the claims unconditional for the resulting decomposition. revision: yes

  2. Referee: [Identities and lower bounds] The lower bounds obtained from regulator positivity and the subsequent balancing problem both presuppose that the gap sequence is intrinsic to the chain. If different decompositions of the same chain produce different gap sequences, the identities and the optimization objective become decomposition-dependent rather than chain-dependent, weakening the comparison claim for chains of equal length and target.

    Authors: This observation is correct and follows from the preceding point. Once the canonical selection rule is fixed, each chain determines a unique gap sequence, so the identities, lower bounds, and balancing objective become intrinsic to the chain. The revised text will state this explicitly and update the comparison claim accordingly. revision: yes

Circularity Check

0 steps flagged

No circularity: identities derived directly from decomposition

full rationale

The paper introduces a decomposition of generators into determiner and regulator (gap), then states explicit identities for sums and reciprocal sums as following from that decomposition and positivity. No equations reduce a claimed prediction or result to a fitted parameter, self-citation, or input by construction. The derivation chain remains self-contained against the stated assumptions; the skeptic concern about non-canonical choice in multi-representation chains is an assumption-validity issue, not a circularity reduction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The paper introduces a new decomposition viewpoint on top of the standard definition of addition chains; the only explicit assumption highlighted is positivity of regulators for bounds. No free parameters or invented physical entities are mentioned.

axioms (1)
  • domain assumption Regulators (gaps) are positive for every addition chain
    Invoked to obtain lower bounds on the sums from the identities.
invented entities (1)
  • Determiner-regulator decomposition of generators no independent evidence
    purpose: To express aggregate statistics of the chain in terms of the gap sequence
    New viewpoint introduced to derive the identities; no independent evidence outside the paper is provided.

pith-pipeline@v0.9.0 · 5675 in / 1343 out tokens · 30819 ms · 2026-05-24T15:45:19.517442+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

10 extracted references · 10 canonical work pages

  1. [1]

    Brauer, On addition chains, Bulletin of the American mathematical Society, vol

    A. Brauer, On addition chains, Bulletin of the American mathematical Society, vol. 45:10, 1939, 736--739

    work page 1939

  2. [2]

    Clift, Calculating optimal addition chains, Computing, vol

    M. Clift, Calculating optimal addition chains, Computing, vol. 91:3, Springer, 1965, pp 265--284

    work page 1965

  3. [3]

    Yang, Xun Qian A Note on 4n=1x+1y+1z, Proceedings of the American Mathematical society, JSTOR, 1982, pp 496--498

    work page 1982

  4. [4]

    46:2, Elsevier, 1994, pp 123--136

    Sander, JW On 4/n= 1/x+ 1/y+ 1/z and Iwaniec′ Half Dimensional Sieve, Journal of Number Theory, vol. 46:2, Elsevier, 1994, pp 123--136

    work page 1994

  5. [5]

    Montgomery, H.L, and Vaughan, R.C, Multiplicative number theory 1:Classical

    work page

  6. [6]

    Robbins, Herbert, A remark on Stirling's formula, Amer. Math. Mon., vol. 62.1, 1955, pp.26--29

    work page 1955

  7. [7]

    71, Siam, 2000

    Meyer, Carl D, Matrix analysis and applied linear algebra vol. 71, Siam, 2000

    work page 2000

  8. [8]

    Nathanson, M.B, Graduate Texts in Mathematics,

    work page

  9. [9]

    3, Springer, 2005

    Roman, Steven and Axler, S and Gehring, FW, Matrix analysis and applied linear algebra, vol. 3, Springer, 2005

    work page 2005

  10. [10]

    R. A. DeVore, Approximation of functions,

    work page