aDarXivDesk
ExploreDocs

General Mathematics

4,924 papers in this slice of arXiv.

All fieldsArtificial IntelligenceMachine LearningComputation and LanguageComputer Vision and Pattern RecognitionNeural and Evolutionary ComputingRoboticsInformation RetrievalHuman-Computer InteractionCryptography and SecurityData Structures and AlgorithmsSoftware EngineeringDistributed, Parallel, and Cluster ComputingProgramming LanguagesSystems and Control
2608.12991
2 days ago

σσσ-Irregularity of Trees with Prescribed Maximum Degree: A Majorization--Duality--Stability Framework

Jasem Hamoud, Duaa Abdullah

Trees of maximum σσσ-irregularity subject to a prescribed maximum degree ΔΔΔ have been characterized for Δ=4Δ=4Δ=4

PreviousNext
and
Δ=5Δ=5Δ=5
, with the extension to arbitrary
Δ⩾3Δ\geqslant 3Δ⩾3
. This paper develops a unified framework for this class of problems built on three pillars. A majorization-theoretic reformulation that decomposes
σσσ
-extremization into a Schur-convex optimization over degree sequences and a rearrangement type optimization over tree realizations, valid for every
Δ⩾3Δ\geqslant 3Δ⩾3
. We establish a duality between the maximization and minimization problems, alongside the general
ΔΔΔ
closed form for
σmax⁡(n,Δ)σ_{\max}(n,Δ)σmax​(n,Δ)
. A stability result showing that the optimality gap between extremal and near extremal trees is bounded independently of
nnn
and grows as
Θ(Δ3)Θ(Δ^3)Θ(Δ3)
.
General Mathematics
2608.12418
3 days ago

Reciprocal Cost on the Positive Rationals

Jonathan Washburn, Sebastian Pardo-Guerra, Milan Zlatanović

We study nonnegative solutions of the reciprocal cost law on the positive rationals and determine which of them admit regular extensions to the positive reals. Using the substitution H=1+FH=1+FH=1+F, we reduce the reciprocal composition law to d'Alembert's functional equation. We prove that every nonnegative solution on Q>0\mathbb{Q}_{>0}Q>0​ is determined by one real weight αpα_pαp​ for each prime ppp, with only a global sign identification. Thus the rational solution space is infinite dimensional. We prove that every nonnegative rational solution has an algebraic extension to R>0\mathbb{R}_{>0}R>0​, but a regular extension exists exactly when the prime weights satisfy αp=λlog⁡pα_p=λ\log pαp​=λlogp for some λ∈Rλ\in\mathbb{R}λ∈R. In this case the extension is unique and belongs to the one-parameter family Fλ(x)=cosh⁡(λlog⁡x)−1F_λ(x)=\cosh(λ\log x)-1Fλ​(x)=cosh(λlogx)−1. Otherwise the rational solution is unbounded on every nonempty open subset of Q>0\mathbb{Q}_{>0}Q>0​, and the regular locus is closed and nowhere dense. We also extend the result to arbitrary nontrivial subgroups of R>0\mathbb{R}_{>0}R>0​, where the alternative is governed by rational rank. Finally, we show that unit logarithmic curvature selects the canonical reciprocal cost J(x)=(x+x−1)/2−1J(x)=(x+x^{-1})/2-1J(x)=(x+x−1)/2−1, while, for a carrier GGG whose logarithm is dense in R\mathbb{R}R, the single asymptotic condition F(es)∼s2/2F(e^s)\sim s^2/2F(es)∼s2/2 implies both regularity and calibration.

General Mathematics
2608.12402
5 days ago

"Phase Transition" in fractional differential equations

Pavel B. Dubovski, Jeffrey A. Slepoi

The basic result in the theory of linear ODEs is that the dimension of the fundamental set of linearly independent solutions is equal to the order of equation. This view is typically extended from integer-order to fractional differential equations. The purpose of this research is to demonstrate that this expectation is incorrect and explain why. We show that linear fractional differential equations may admit additional linearly independent solutions beyond their order. This phenomenon occurs when the coefficients of the equation cross certain threshold values. This phase transition produces a hyper-dimensional fundamental set of solutions and offers new insight into fractional differential equations and leads to the revisions of the statements of initial and boundary value problems.

General Mathematics
2608.11257
8 days ago

Combinatorial and Sampling Approaches to Determining the Priority Vector from an Incomplete Pairwise Comparison Matrix on the Set of Spanning Trees

Vitaliy Tsyganok, Oksana Mulesa

This paper considers combinatorial and sampling approaches to deriving a priority vector from an incomplete pairwise comparison matrix. It is shown that the informationally significant combinations of matrix elements are in one-to-one correspondence with spanning trees of the graph induced by the matrix. The combinatorial approach is based on the enumeration of all spanning trees and the componentwise aggregation of the priority vectors derived from them. To reduce computational complexity, a sampling Monte Carlo approach with a sequential stopping criterion is proposed, ensuring a prescribed accuracy for estimating the components of the resulting vector.

General Mathematics
2608.06442
9 days ago

Asymptotic Formulae For Reciprocal Partitions

Nilotpal Kanti Sinha

For integers 1≤m≤n1\le m\le n1≤m≤n, let s(n,m)s(n,m)s(n,m) denote the number of mmm-tuples (k1,…,km)(k_1,\ldots,k_m)(k1​,…,km​) of nonnegative integers satisfying n=∑j=1mkjj.n=\sum_{j=1}^{m}\frac{k_j}{j}.n=j=1∑m​jkj​​. We obtain a complete asymptotic expansion for log⁡s(n,m)\log s(n,m)logs(n,m), uniformly for all n≥mn\ge mn≥m as m→∞m\to\inftym→∞. The coefficients are given explicitly in terms of limiting prime-block functions arising from the residue structure of the problem. We also determine the full hierarchy of multiplicative corrections in the sparse regime, identifying explicit constants at every fixed order and, in particular, the first correction constants 1/41/41/4 and (1−log⁡2)/4(1-\log 2)/4(1−log2)/4. The results give uniform two-parameter asymptotics as both the target and the number of allowed reciprocal parts grow.

General Mathematics
2608.06438
9 days ago

Can πππ generate itself? A Monte Carlo analysis of 314 trillion digits

Alessandro Razeto, Nicola Rossi

At the end of 2025, a record computation of πππ reached 314 trillion decimal digits, providing the largest numerical dataset ever generated for this constant. We exploit this unprecedented dataset to investigate whether the digits of πππ themselves can serve as a source of pseudorandom numbers for estimating πππ through the simplest Monte Carlo method. Our results go beyond the normality hypothesis by providing empirical evidence of a high degree of statistical randomness in the available digits, although not of digit independence, which cannot hold for a deterministic sequence. By optimizing the mapping of the digit sequence into Monte Carlo samples, we obtain the highest precision allowed by the dataset. As predicted, the method successfully reproduces the first sequence of decimal digits, demonstrating that the largest available dataset of πππ digits can be used to recover π≈3.141593π\approx 3.141593π≈3.141593 through Monte Carlo simulation.

General Mathematics
2608.06436
9 days ago

Crossed Homomorphisms on associative superalgebras

RB Yadav, Arpan Sharma

In this paper, we introduce the notion of crossed homomorphisms on associative superalgebras. We show that crossed homomorphisms are precisely the Maurer--Cartan elements of a naturally associated graded Lie algebra. This characterization enables us to construct a cohomology theory of crossed homomorphisms. We then investigate formal deformations of crossed homomorphisms and derive the corresponding deformation equations. It is proved that the infinitesimal part of a formal deformation is a 1-cocycle in the associated cohomology, while the obstruction to extending a deformation of finite order is a 2-cocycle.

General Mathematics
2608.06435
9 days ago

Chebyshev Bias for Largest Prime Factors

Nilotpal Kanti Sinha

Let P+(n)P^+(n)P+(n) be the largest prime factor of nnn, and let χ=χ−4χ=χ_{-4}χ=χ−4​ be 111 on primes 1(mod4)1\pmod41(mod4) and −1-1−1 on primes 3(mod4)3\pmod43(mod4). For fixed k≥2k\ge2k≥2 we study Dk(x)=∑n≤xΩ(n)=kχ(P+(n)).D_k(x)=\sum_{\substack{n\le x\\ Ω(n)=k}}χ(P^+(n)).Dk​(x)=n≤xΩ(n)=k​∑​χ(P+(n)). Thus Dk(x)D_k(x)Dk​(x) compares the two residue classes according to the largest prime factor of integers having exactly kkk prime factors, counted with multiplicity. Assuming RH for ζ(s)ζ(s)ζ(s) and L(s,χ)=β(s)L(s,χ)=β(s)L(s,χ)=β(s), we prove a pointwise explicit formula. The main term is a fixed negative contribution plus an absolutely convergent oscillating sum over the zeros ρ=12+iγρ=\frac12+iγρ=21​+iγ of L(s,χ)L(s,χ)L(s,χ). The ordering of the prime factors produces kkk Perron denominators, and the principal coefficient of a zero is Ok((1+∣γ∣)−k)O_k((1+|γ|)^{-k})Ok​((1+∣γ∣)−k). We then show that the total size of all zero terms is strictly smaller than the fixed contribution. Hence Dk(x)<0D_k(x)<0Dk​(x)<0 for all sufficiently large xxx. The analogous problem without fixing kkk is still open.

General Mathematics
2608.05194
12 days ago

Non-Archimedean Cauchy-Schwarz Angle-Length and Chebyshev Arithmetic Mean Inequalities

K. Mahesh Krishna

Let K\mathbb{K}K be a non-Archimedean valued field. Let n∈Nn \in \mathbb{N}n∈N. For every (aj)j=1n,(bj)j=1n∈Kn(a_j)_{j=1}^n, (b_j)_{j=1}^n \in \mathbb{K}^n(aj​)j=1n​,(bj​)j=1n​∈Kn, we show that align* |∑_j=1ⁿa_jb_j|²≤ {|∑_j=1ⁿ a_j²||∑_k=1ⁿb²_k|, ₁≤ j<k ≤ n|a_jb_k-a_kb_j|²}, align* align* |∑_j=1ⁿ a_j²||∑_k=1ⁿb²_k|≤ {|∑_j=1ⁿ a_jb_j|², ₁≤ j<k ≤ n|a_jb_k-a_kb_j|²}, align* align* |AM(a_j)_j=1ⁿ||AM(b_j)_j=1ⁿ|≤ { |AM(a_jb_j)_j=1ⁿ|, (1/|n|²)₁≤ j < k ≤ n|a_j-a_k||b_j-b_k|}, align* align* |AM(a_jb_j)_j=1ⁿ|≤ { |AM(a_j)_j=1ⁿ||AM(b_j)_j=1ⁿ|, (1/|n|²)₁≤ j < k ≤ n|a_j-a_k||b_j-b_k|}, align* where AMAMAM denotes the arithmetic mean. First and second are non-Archimedean versions of Cauchy-Schwarz angle-length and third and fourth are Chebyshev arithmetic mean inequalities. Unlike in the Archimedean case, no conditions are required to derive non-Archimedean Chebyshev arithmetic mean inequalities.

General Mathematics
2608.05193
13 days ago

Infinite and finite series involving central binomial coefficients and closed forms of generalized hypergeometric functions

Ganesh Bahadur Basnet, Narayan Prasad Pahari, Feng Qi +1

Let Z−=∖{−1,−2,… }\mathbb{Z}^-=\setminus\{-1,-2,\dotsc\}Z−=∖{−1,−2,…}. In 2023, Qi and Lim gave two claims for summing the infinite series ∑k=1∞(2kk)1α+k(±14)k,α∈C∖Z−.\sum_{k=1}^{\infty} \binom{2k}{k} \frac{1}{α+k} \biggl(\frac{\pm1}{4}\biggr)^k, \quad α\in\mathbb{C}\setminus\mathbb{Z}^-.k=1∑∞​(k2k​)α+k1​(4±1​)k,α∈C∖Z−. In present paper, the authors establish several sum functions of the infinite and finite series ∑k=1∞(2kk)1α+k(z4)kand∑k=1n(2kk)1α+k(z4)k\sum_{k=1}^{\infty}\binom{2k}{k}\frac{1}{α+k}\biggl(\frac{z}{4}\biggr)^k \quad\text{and}\quad \sum_{k=1}^{n}\binom{2k}{k}\frac{1}{α+k}\biggl(\frac{z}{4}\biggr)^kk=1∑∞​(k2k​)α+k1​(4z​)kandk=1∑n​(k2k​)α+k1​(4z​)k for α∈C∖Z−α\in\mathbb{C}\setminus\mathbb{Z}^-α∈C∖Z− and n∈N={1,2,… }n\in\mathbb{N}=\{1,2,\dotsc\}n∈N={1,2,…} in terms of the Gauss hypergeometric functions 2F1{}_2F_12​F1​ and the generalized hypergeometric functions 3F2{}_3F_23​F2​ for α∈C∖Z−α\in\mathbb{C}\setminus\mathbb{Z}^-α∈C∖Z− and n∈Nn\in\mathbb{N}n∈N. In light of the Euler integral representation of the Gauss hypergeometric function 2F1{}_2F_12​F1​, the author present several closed forms of two Gauss hypergeometric functions 2F1{}_2F_12​F1​, two generalized hypergeometric functions 3F2{}_3F_23​F2​, and the classical incomplete beta functions Bz(12,12+n)B_z\bigl(\frac12, \frac{1}{2}+n\bigr)Bz​(21​,21​+n) and Bz(12,1+n)B_z\bigl(\frac12, 1+n\bigr)Bz​(21​,1+n). With the help of the Euler hypergeometric transform, the authors derive closed forms of five Gauss hypergeometric functions. In addition, the authors also obtain a closed form of the differential operator [(1−z)d⁡d⁡z(1−z)]narcsin⁡zz(1−z)\bigl[(1-z)\frac{\operatorname{d}}{\operatorname{d}z}(1-z)\bigr]^n \frac{\arcsin\sqrt{z}}{\sqrt{z(1-z)}}[(1−z)dzd​(1−z)]nz(1−z)​arcsinz​​ for n∈N0={0}∪Nn\in\mathbb{N}_0=\{0\}\cup\mathbb{N}n∈N0​={0}∪N.

General Mathematics
2608.05192
13 days ago

Extensions of several famous combinatorial identities via hypergeometric functions

Arjun Kumar Rathie, Feng Qi, Dongkyu Lim

The objective of this paper is to develop an extension of Kummer's second theorem and to establish generalized forms of four classical combinatorial identities---Knuth's old sum (also known as Reed--Dawson's combinatorial identity), Riordan's combinatorial identity, Gould's combinatorial identity, and Touchard's combinatorial identity---using a hypergeometric-series approach. Several new identities also arise as special cases of our main results.

General Mathematics
2608.05190
13 days ago

Some Generalizations of the Bridge and Torch Problem

Pang Ern Thang, Gerard Sayson

For the classic bridge and torch problem with crossing times {1,…,n}\left\{1,\ldots,n\right\}{1,…,n}, we derive a closed-form expression for the optimal crossing time T(n)T\left(n\right)T(n) using a recurrence relation derived from the problem's optimal substructure, thus obtaining T(n)=n24+3n−5+(−1)n−18T\left(n\right)=\frac{n^2}{4}+3n-5+\frac{\left(-1\right)^n-1}{8}T(n)=4n2​+3n−5+8(−1)n−1​ which holds for all n≥2n\ge 2n≥2. We generalize the problem to a bridge of capacity 3 and obtain the optimal crossing time T3(n)=n26+2n−18136+(−1)n4−29cos⁡(2nπ3)T_3\left(n\right)=\frac{n^{2}}{6}+2n-\frac{181}{36}+\frac{\left(-1\right)^{n}}{4}-\frac{2}{9}\cos\left(\frac{2nπ}{3}\right)T3​(n)=6n2​+2n−36181​+4(−1)n​−92​cos(32nπ​) which holds for all n≥7n\ge 7n≥7. As such, we obtain a new sequence A392834 in the On-Line Encyclopedia of Integer Sequences. Lastly, we also explore this problem for star graphs, and see how we can recover some classic identities involving the sum of floor functions.

General Mathematics
2608.06403
13 days ago

From Schwartz Space to Structural Aging (Towards a small history of Old Age mathematics)

Daniel Parrochia

Starting with some historical considerations, we examine the various attempts to model old age that fall within the realm of mathematical physics. For example, we study the mathematical theory of declining functions, which, since Laurent Schwartz's work on distributions, are generally defined on what is called a "Schwartz space." We explain how this essentially analytic formalization applies to the representation of the decline of vitality (or of a number of vital functions). We then show that there exist many other possible formalizations that utilize the resources of algebra, in particular the theory of "inclines" of Cao, Kim, and Roush and that of relational structures, on which an algebraic notion of "age," in Cameron's sense, can be defined. But we maintain, in conclusion, that a more positive representation of old age, consistent, moreover, with what a number of cultural sources from Antiquity to the 20th century tell us, should mobilize another type of formalization, capable of selecting, within the overall debate between the living being and its environment, persistent structures that do not decline amidst the decline of others.

General Mathematics
2608.05189
15 days ago

The Gregory function and its completed Gregory transform

Grant Molnar

We study the entire interpolation G(z)=∫01(xz) dx\mathcal{G}(z)=\int_0^1 \binom{x}{z}\,dxG(z)=∫01​(zx​)dx of the Gregory coefficients. Its completion satisfies the positive Markov-transform identity πzsin⁡(πz)G(z)=∑n=1∞n∣Gn∣n−z.\frac{πz}{\sin(πz)}\mathcal{G}(z) =\sum_{n=1}^{\infty}\frac{n\left|G_n\right|}{n-z}.sin(πz)πz​G(z)=n=1∑∞​n−zn∣Gn​∣​. Consequently, every zero is real and simple; the negative zeros are the integers −1,−2,…-1,-2,\ldots−1,−2,…, and one zero ρnρ_nρn​ lies in each (n,n+1)(n,n+1)(n,n+1). We derive complete logarithmic asymptotics for ρn−nρ_n-nρn​−n, determine the Cartwright growth and canonical products of G\mathcal{G}G, and realize 1/ρn1/ρ_n1/ρn​ spectrally. The resulting relative determinant yields γ=∑n=1∞(1n−1ρn).γ=\sum_{n=1}^{\infty}\left(\frac{1}{n}-\frac{1}{ρ_n}\right).γ=n=1∑∞​(n1​−ρn​1​).

General Mathematics
2608.12384
15 days ago

Multi-point variants of the Newton-Raphson-Simpson method arising from organizing a formal zero according to a function φφφ

Mario DeFranco

Fix an integer L≥1L \geq 1L≥1, and a function φ ⁣:Z≥1→[0,L]∩Zφ\colon \mathbb{Z}_{\geq 1} \rightarrow [0,L] \cap \mathbb{Z}φ:Z≥1​→[0,L]∩Z with φ−1({0})={1}φ^{-1}(\{0\}) = \{ 1\}φ−1({0})={1}. We define a multi-point variant of the Newton-Raphson-Simpson, which we call the max-phi method, as follows. We use φφφ to define the iteration number of a rooted plane tree. Then we construct formal series that are weighted generating functions of rooted plane trees with iteration number at most NNN. Finally we use these formulas to define the max-phi method applied to an arbitrary LLL-differentiable function.

General Mathematics
2608.00092
16 days ago

Explicit Green's Functions and Adjoint Problems for Differential Equations with Linear Functional Perturbations

Alberto Cabada, Paula Cambeses-Franco, Lucía López-Somoza

In this paper we study a functional differential equation subject to two-point boundary value conditions, where the functional dependence is introduced through an operator of the form equation* ∑_k=1^lγ_k(t)C_k(u), equation* where Ck:C(I)→R\mathcal{C}_{k}:C(I) \rightarrow \mathbb{R}Ck​:C(I)→R, k=1,…lk=1, \ldots lk=1,…l, (I:=[a,b]I:=[a,b]I:=[a,b]) are linear continuous operators and γk∈L1(I)γ_{k} \in \mathcal{L}^{1}(I)γk​∈L1(I) for all k=1,…,lk=1, \ldots, lk=1,…,l. This formulation encompasses, among others, equations with piecewise constant arguments and those with integral-type dependence. We analyze this class of equations by deducing and characterizing their Green's function, as well as by computing the related adjoint problem. This approach enables us to establish connections among different types of functional equations and to relate equations with piecewise constant arguments to impulsive differential equations and non local boundary value problems. Next, we develop a series of comparison principles and results that allow us to characterize the regions where the Green's function of the original problem and of its adjoint maintain a constant sign. Finally, we illustrate the theoretical finding with representative examples.

General Mathematics
2607.27503
17 days ago

Character Fourier Spectra of Circular Units and Twisted Bernoulli Class Components

Peter Chocian

Let χχχ be a primitive odd Dirichlet character of conductor fff. For the universal projector polynomials PmP_mPm​ introduced in arXiv:2607.23177, defined by ∑mPm(X)Ym=−log⁡(1−X(1−e−Y))\sum_m P_m(X)Y^m = -\log(1-X(1-e^{-Y}))∑m​Pm​(X)Ym=−log(1−X(1−e−Y)), we evaluate the character Fourier spectrum at ht=ζft/(ζft−1)h_t = ζ_f^t/(ζ_f^t-1)ht​=ζft​/(ζft​−1): for every odd mmm, ∑tχˉ(t)Pm(ht)=τ(χˉ)Bm,χ/(m m!)\sum_t \barχ(t)P_m(h_t) = τ(\barχ)B_{m,χ}/(m\,m!)∑t​χˉ​(t)Pm​(ht​)=τ(χˉ​)Bm,χ​/(mm!). After reduction at any prime above p∤fp \nmid fp∤f, the case m=p−jm = p-jm=p−j identifies this spectrum, including its exact nonzero scalar, with the divided generalized Bernoulli value attached to χω−jχω^{-j}χω−j; the spectral-zero and Bernoulli-zero criteria therefore agree over every residue field, with no splitting hypothesis on the coefficient field. We connect the identity with the local Kummer spectrum of the circular unit 1−ζfζp1-ζ_fζ_p1−ζf​ζp​ and carry the programme through in the first non-real case: for the two primitive quartic characters modulo 5 and primes p<500p < 500p<500, p≡1(mod20)p \equiv 1 \pmod{20}p≡1(mod20), exactly eleven zero lines occur. On each line an integral character projection of 1−ζ5ζp1-ζ_5ζ_p1−ζ5​ζp​ is everywhere locally unramified, a finite split-prime Artin computation proves it is not a global ppp-th power, and the character-wise Main Conjecture shows the radical generates the complete order-ppp Hilbert class component. Six of the eleven components occur at classically regular primes. At p=61p = 61p=61 the two conjugate characters contribute on different indices. A deterministic integer-arithmetic program (ancillary file) verifies the enumeration, the divided digits, and every certificate.

General Mathematics
2608.05186
17 days ago

Incidence-based Combinatorial Geometry on Cell Complexes

Andrey P. Jivkov, Muhammad Azeem, Aneirin Griffiths +1

Combinatorial Mesh Calculus (CMC) formulates conservation laws directly on cell complexes using combinatorial differential forms and their cochain representations. This note develops an incidence-based geometric extension of that framework for directional quantities and local geometric structure on explicit cellular organisation. The central construction is a family of local fibres generated intrinsically by the incidence structure of the complex. Each vertex carries a vector space spanned by its incident edge directions, providing a combinatorial analogue of a tangent space whose dimension reflects local topology. These fibres define local coefficient spaces for incidence-based vector-, covector- and endomorphism-valued cochains. Directed transport maps compare states attached to neighbouring fibres, while a canonical solder form relates fibre directions to the underlying cell-complex structure. Together they give rise to finite combinatorial analogues of transport, torsion, curvature and metric structure. Evaluation cup products provide pairings between kinematic and force-like quantities, while bundle Hodge operators induced by the fibre metric relate vector- and covector-valued cochains, and weighted covariant incidence operators define degree-raising transport-corrected operations on incidence cochains. The framework does not assume a smooth manifold, fixed-rank bundle, cellular sheaf or local system. Instead, geometric structure is generated directly from the organisation represented by the cell complex. The resulting theory establishes a combinatorial scaffold for geometry on explicit cellular organisation and identifies the principal mathematical questions required for its further development, including admissible transport classes, metric compatibility, Cartan-type structure equations and locality-dependent algebraic structures.

General Mathematics
2608.00070
17 days ago

Asymptotic Analysis of Nonsymmetric Gaussian and Archimedean Compound Means

Tomislav Burić, Lenka Mihoković, Toni Milas

This paper studies asymptotic expansions of the Gaussian and Archimedean compounds of two arbitrary bivariate means that need not be symmetric. Recursive algorithms for the coefficients of such expansions were previously obtained only in the symmetric case. Here, this restriction is removed and recursions are derived for the general nonsymmetric setting. All coefficient calculations are carried out in the algebra of formal power series. This introduces an additional case in the recursion, as well as singular configurations in which the formal invariance equation does not determine the coefficients recursively. The formal problem is kept separate from the existence and convergence of the iterative procedures. We also connect the first formal coefficients with Farhi's metric, recall its known global convergence criterion, and derive the corresponding local rates. The algorithms are illustrated using weighted power means and several known identities. Finally, the method is extended to expansions with sign-dependent coefficients and is applied to the neo-Pythagorean means and to a recently introduced two-parameter family.

General Mathematics
2608.00067
17 days ago

On non-symmetric ttt-convexity

Paolo Leonetti

Zsolt Páles asked whether there exists a function f:R→Rf: \mathbb{R}\to \mathbb{R}f:R→R such that f(x+2y3)≤f(x)+2f(y)3 for all x≤yf\left(\frac{x+2y}{3}\right) \le \frac{f(x)+2f(y)}{3} \quad \text{ for all }x\le yf(3x+2y​)≤3f(x)+2f(y)​ for all x≤y which is not midconvex. We show that the answer is negative.

General Mathematics