Period (algebraic geometry)

A real number is called a period if it is the difference of volumes of regions of Euclidean space given by polynomial inequalities with rational coefficients. More generally a complex number is called a period if its real and imaginary parts are periods.

Periods are numbers that arise as integrals of algebraic functions over domains that are described by algebraic equations or by inequalities with rational coefficients (Weisstein 2019). Periods can be defined as complex numbers whose real and imaginary parts are values of absolutely convergent integrals of rational functions with rational coefficients, over domains in ${\displaystyle \mathbb {R} ^{n}}$ given by polynomial inequalities with rational coefficients (Kontsevich & Zagier 2001, p. 3). The coefficients of the rational functions and polynomials can be generalised to algebraic numbers since integrals and irrational algebraic numbers are expressible in terms of areas of suitable domains.

Besides the algebraic numbers, the following numbers are known to be periods:

• The natural logarithm of any positive algebraic number a, which is ${\displaystyle \int _{1}^{a}{\frac {1}{x}}\ \mathrm {d} x}$
• π${\displaystyle =\int _{0}^{1}{\frac {4}{x^{2}+1}}\ \mathrm {d} x}$
• Elliptic integrals with rational arguments
• All zeta constants (the Riemann zeta function of an integer) and multiple zeta values
• Special values of hypergeometric functions at algebraic arguments
• Γ(p/q)^q for natural numbers p and q.

An example of a real number that is not a period is given by Chaitin's constant Ω. Any other non-computable number also gives an example of a real number that is not a period. Currently there are no natural examples of computable numbers that have been proved not to be periods, however it is possible to construct artificial examples (Yoshinaga 2008). Plausible candidates for numbers that are not periods include e, 1/π, and Euler–Mascheroni constant γ.

The periods are intended to bridge the gap between the algebraic numbers and the transcendental numbers. The class of algebraic numbers is too narrow to include many common mathematical constants, while the set of transcendental numbers is not countable, and its members are not generally computable.

The set of all periods is countable, and all periods are computable (Tent 2010), and in particular definable.

Many of the constants known to be periods are also given by integrals of transcendental functions. Kontsevich and Zagier note that there "seems to be no universal rule explaining why certain infinite sums or integrals of transcendental functions are periods".

Kontsevich and Zagier conjectured that, if a period is given by two different integrals, then each integral can be transformed into the other using only the linearity of integrals, changes of variables, and the Newton–Leibniz formula

${\displaystyle \int _{a}^{b}f'(x)\,dx=f(b)-f(a)}$

(or, more generally, the Stokes formula).

A useful property of algebraic numbers is that equality between two algebraic expressions can be determined algorithmically. The conjecture of Kontsevich and Zagier would imply that equality of periods is also decidable: inequality of computable reals is known recursively enumerable; and conversely if two integrals agree, then an algorithm could confirm so by trying all possible ways to transform one of them into the other one.

It is not expected that Euler's number e and Euler–Mascheroni constant γ are periods. The periods can be extended to exponential periods by permitting the product of an algebraic function and the exponential function of an algebraic function as an integrand. This extension includes all algebraic powers of e, the gamma function of rational arguments, and values of Bessel functions. If, further, Euler's constant γ is added as a new period, then according to Kontsevich and Zagier "all classical constants are periods in the appropriate sense".

• Jacobian variety
• Gauss–Manin connection
• Mixed motives (math)
• Tannakian formalism

• Belkale, Prakash; Brosnan, Patrick (2003), "Periods and Igusa local zeta functions", International Mathematics Research Notices, 2003 (49): 2655–2670, doi:10.1155/S107379280313142X, ISSN 1073-7928, MR 2012522
• Kontsevich, Maxim; Zagier, Don (2001), "Periods" (PDF), in Engquist, Björn; Schmid, Wilfried (eds.), Mathematics unlimited—2001 and beyond, Berlin, New York: Springer-Verlag, pp. 771–808, ISBN 978-3-540-66913-5, MR 1852188
• Waldschmidt, Michel (2006), "Transcendence of periods: the state of the art" (PDF), Pure and Applied Mathematics Quarterly, 2 (2): 435–463, doi:10.4310/PAMQ.2006.v2.n2.a3, ISSN 1558-8599, MR 2251476
• Tent, Katrin; Ziegler, Martin (2010), "Computable functions of reals" (PDF), Münster Journal of Mathematics, 3: 43–66
• Weisstein, Eric W. "Periods". mathworld.wolfram.com. Retrieved 2019-06-19.
• Yoshinaga, Masahiko (2008-05-03). "Periods and elementary real numbers". arXiv:0805.0349 [math.AG].CS1 maint: ref=harv (link)

• PlanetMath: Period