Fact-checked by Grok 2 weeks ago

Sum of squares function

In , the sum of squares function, commonly denoted r_k(n), quantifies the number of ways to express a positive n as the sum of k squares of integers, where representations distinguish order, signs, and zeros (i.e., solutions to n = x_1^2 + x_2^2 + \dots + x_k^2 with x_i \in \mathbb{Z}). This function arises in the analytic theory of quadratic forms and plays a central role in understanding Diophantine equations of the form \sum x_i^2 = n. For k = 2, on sums of two squares characterizes which n admit representations: an odd prime p can be written as p = a^2 + b^2 with a, b > 0 p \equiv 1 \pmod{4}, and more generally, n is a sum of two squares precisely when every prime congruent to $3 \pmod{4} in its factorization has even exponent. The value of r_2(n) is then given by the formula r_2(n) = 4(d_1(n) - d_3(n)), where d_1(n) and d_3(n) are the numbers of divisors of n congruent to $1 \pmod{4} and $3 \pmod{4}, respectively; equivalently, for n = 2^c \prod p_i^{a_i} \prod q_j^{b_j} with p_i \equiv 1 \pmod{4} and q_j \equiv 3 \pmod{4}, r_2(n) = 4 \prod (a_i + 1) if all b_j are even, and $0 otherwise. Jacobi extended this to higher k, providing explicit formulas such as r_4(n) = 8 \sum_{d|n, 4 \nmid d} d for odd n, reflecting the multiplicative structure of the function. Lagrange's four-square theorem establishes that every positive integer n satisfies r_4(n) > 0, meaning all naturals are sums of at most four squares, with the theorem proved in 1770. For larger k, r_k(n) grows asymptotically like c_k n^{(k/2)-1}, linking to broader problems in additive number theory such as Waring's problem, where the minimal g(2) = 4 confirms four squares suffice globally. These results underscore the function's connections to modular forms, theta functions, and the distribution of lattice points on spheres.

Definition and Basics

Definition

The sum of squares function, commonly denoted r_k(n), quantifies the number of ways a positive n can be expressed as the sum of k squares of . Specifically, it counts the solutions (x_1, x_2, \dots, x_k) \in \mathbb{Z}^k to the x_1^2 + x_2^2 + \dots + x_k^2 = n, where zeros are permitted, negative are distinguished from positives (i.e., (-x_i)^2 = x_i^2 but the tuples are counted separately if signs differ), and the order of the summands matters, so permutations of the x_i yield distinct representations. This convention aligns with the function's role in , where it facilitates the study of quadratic forms and modular functions. For instance, r_2(1) = 4 corresponds to the representations (\pm 1, 0) and (0, \pm 1). The function originates from classical problems in number theory concerning the representability of integers by sums of squares, with foundational analytic expressions derived by C. G. J. Jacobi in 1829 for even values of k up to 8. Jacobi's work in Fundamenta nova theoriae functionum ellipticarum linked r_k(n) to elliptic theta functions, establishing the generating function \sum_{n=0}^\infty r_k(n) q^n = \vartheta_3(q)^k, where \vartheta_3(q) = \sum_{m=-\infty}^\infty q^{m^2} is the Jacobi theta function of the third kind. This theta-series representation underscores the function's deep connections to modular forms and lattice theory, though the basic definition remains independent of these advanced structures. In standard texts, such as Hardy and Wright's An Introduction to the Theory of Numbers, the function is presented with this full counting convention to capture all integral representations without restrictions on primitivity or positivity.

Notation and Conventions

The sum of squares function is standardly denoted by r_k(n), where k is a positive integer and n is a positive integer, representing the number of ways to express n as the sum of k squares of integers, that is, the number of solutions in integers x_1, x_2, \dots, x_k \in \mathbb{Z} to the equation x_1^2 + x_2^2 + \dots + x_k^2 = n. This notation counts all ordered k-tuples, distinguishing different orders of the summands as distinct representations; it also allows zeros and distinguishes signs, so that, for example, (-x_i)^2 = x_i^2 but the signed tuples are treated separately in the counting. For the specific case k=2, the function is often abbreviated as r(n) or r_2(n). An illustrative example is r_2(5) = 8, arising from the ordered pairs (\pm 1, \pm 2) and (\pm 2, \pm 1) with all four sign combinations for each permutation. Similarly, zeros are permitted, as in r_3(4) = 6, which includes representations like (\pm 2, 0, 0) and permutations. While this ordered, signed convention with zeros is the predominant one in , variations appear in some contexts; for instance, the number of representations using only non-negative integers (disregarding order and signs) is sometimes denoted differently, such as \sigma_k(n) or studied via unordered partitions into squares, but these are not standard for r_k(n). The generating function for r_k(n) is the k-th power of the Jacobi \theta_3(q) = \sum_{m=-\infty}^{\infty} q^{m^2}, reflecting the inclusion of all integers.

Generating Functions

Jacobi Theta Function

The Jacobi theta function, one of the four classical theta functions introduced by , is defined as \vartheta_3(z \mid \tau) = \sum_{n=-\infty}^{\infty} \exp\left(2\pi i n z + \pi i n^2 \tau\right), where \tau lies in the upper half-plane \mathbb{H} and z \in \mathbb{C}. For the purposes of generating functions in , the nome q = e^{\pi i \tau} is often used, simplifying the form to \vartheta_3(0 \mid \tau) = \sum_{n=-\infty}^{\infty} q^{n^2}, which converges absolutely for \operatorname{Im}(\tau) > 0. This function plays a central role as the for the sum of squares function r_k(n), which counts the number of solutions to n = x_1^2 + \cdots + x_k^2. Specifically, the k-th power of the yields \vartheta_3(0 \mid \tau)^k = \sum_{n=0}^{\infty} r_k(n) q^n, where the constant term accounts for the representation of 0, and r_k(n) includes representations allowing zeros, , and . This arises from the product over sums for each square, making \vartheta_3^k the for even k in particular, though it holds formally for any k. The utility of this connection stems from the modular properties of \vartheta_3, which is a modular form of weight $1/2 for the congruence subgroup \Gamma_0(4) with a specific multiplier system. Key transformations include \vartheta_3(\tau + 1) = \vartheta_3(\tau) and \vartheta_3(-1/\tau) = \sqrt{-i\tau} \vartheta_3(\tau), derived via the . These properties enable the extraction of explicit formulas for r_k(n) by expanding \vartheta_3^k in terms of or other basis of modular forms spaces, as Jacobi did for k=2,4,6,8 in his 1829 work on elliptic functions. For instance, for k=4, \vartheta_3^4(\tau) equals \frac{1}{\pi^2} \left( 4 G_2(4\tau) - G_2(\tau) \right), leading to Jacobi's formula r_4(n) = 8 \sum_{d \mid n, \, 4 \nmid d} d. Higher powers \vartheta_3^k for even k remain modular forms of weight k/2, facilitating analytic continuations and identities that reveal structure in r_k(n), such as sums or number relations. This framework, building on Jacobi's foundational contributions, underpins much of the of quadratic forms and has been detailed in standard treatments.

Properties and Expansions

The for the sum-of-squares function r_k(n) is given by the k-th power of the , \theta(\tau)^k = \sum_{n=0}^\infty r_k(n) q^n, where q = e^{\pi i \tau} with \Im(\tau) > 0, and \theta(\tau) = \theta_3(0|\tau) = \sum_{m=-\infty}^\infty q^{m^2}. This function is holomorphic on the upper half-plane \mathbb{H} and serves as a theta series attached to the \mathbb{Z}^k. For even k, \theta(\tau)^k is a of weight k/2 on the \Gamma_0(4) with a specific nebentypus character. A fundamental property arises from the identity, which provides an expansion for the : \theta(\tau) = \prod_{m=1}^\infty (1 - q^{2m})(1 + q^{2m-1})^2. Raising this to the k-th power yields the product form for the , \theta(\tau)^k = \left[ \prod_{m=1}^\infty (1 - q^{2m})(1 + q^{2m-1})^2 \right]^k, facilitating connections to partition theory and elliptic functions. This non-vanishing product in \mathbb{H} underscores the 's role in , ensuring the has no zeros in the upper half-plane. Modular transformation properties are derived from Poisson summation and define the behavior under the action of \mathrm{SL}_2(\mathbb{Z}). Specifically, for the inversion \tau \mapsto -1/\tau, \theta(-1/\tau) = \sqrt{-i\tau} \, \theta(\tau), with \theta(\tau+1) = \theta(\tau), establishing invariance under translation by 1 up to the root of unity factor in the full Jacobi theta. These transformations extend to powers: \theta(\tau)^k transforms with the factor (-i\tau)^{k/2} under inversion, confirming its modular form status for appropriate subgroups. For k=2, this yields Jacobi's identity linking r_2(n) = 4(d_1(n) - d_3(n)), where d_i(n) counts divisors of n congruent to i \pmod{4}, via equating the q-expansion to a twisted . For higher even k, such as k=4, the generating function admits an expansion in terms of Eisenstein series of level 4: \theta(\tau)^4 = \frac{1}{\pi^2} \left( 4 G_2(4\tau) - G_2(\tau) \right), where G_2(\tau) is the weight-2 , leading to Jacobi's four-squares theorem r_4(n) = 8 \sum_{d|n, \, 4 \nmid d} d. Similar Eisenstein expansions hold for k=6,8, expressing \theta(\tau)^k as linear combinations of basis forms in the space of modular forms of weight k/2 on \Gamma_0(4), with dimensions determined by the genus of the modular curve. These identities highlight the arithmetic significance, bounding r_k(n) via Hecke operators and .

Explicit Formulas

For k=2

The explicit formula for the sum of squares function r_2(n), which counts the number of integer solutions to x^2 + y^2 = n (including orders, signs, and zeros), is provided by Jacobi's two-square theorem:
r_2(n) = 4 \left( d_1(n) - d_3(n) \right),
where d_i(n) is the number of positive divisors of n congruent to i modulo 4. This identity, originally established by , quantifies the representations precisely and extends Fermat's earlier result on which primes can be expressed as sums of two squares.
The formula arises from the multiplicative structure of the and the behavior of quadratic residues modulo 4. For a prime p \equiv 3 \pmod{4} raised to an odd power in the prime factorization of n, d_1(n) = d_3(n), yielding r_2(n) = 0; conversely, if all such exponents are even, the difference d_1(n) - d_3(n) equals the product over primes p \equiv 1 \pmod{4} of (e_p + 1), where e_p is the exponent of p, multiplied by 1 for the contribution from powers of 2. Thus, r_2(n) > 0 every prime congruent to 3 4 divides n to an even power, aligning with Fermat's sum-of-two-squares theorem. For example, consider n = 5 = 1^2 + 2^2. The divisors are 1 and 5, both \equiv 1 \pmod{4}, so d_1(5) = 2 and d_3(5) = 0, giving r_2(5) = 8 representations: (\pm 1, \pm 2), (\pm 2, \pm 1). For n = [3](/page/3), a prime \equiv 3 \pmod{4}, the divisors 1 and 3 yield d_1(3) = 1 and d_3(3) = 1, so r_2(3) = 0. These cases illustrate how the formula captures both the existence and multiplicity of representations. The theorem has been proved using various methods, including generating functions via the Jacobi theta function \theta_3(q) = \sum_{m=-\infty}^\infty q^{m^2}, where \theta_3(q)^2 = \sum_{n=0}^\infty r_2(n) q^n, and equating coefficients after expansion. Elementary proofs rely on counting lattice points or factorizations, emphasizing its foundational role in .

For k=3

A n can be expressed as the sum of three squares of integers it is not of the form $4^a (8b + 7) for nonnegative integers a and b; this is , proved using the theory of quadratic forms.90002-5) When such a representation exists, the number r_3(n) of ordered triples (x, y, z) \in \mathbb{Z}^3 satisfying x^2 + y^2 + z^2 = n (counting signs and zeros distinctly) is given by an explicit formula involving Hurwitz class numbers, which generalize ordinary class numbers to account for square factors in the discriminant. The Hurwitz class number H(d) for positive integer d is the weighted sum over equivalence classes of positive definite binary forms of discriminant -4d, where each class is weighted by the reciprocal of the order of its . The formula for r_3(n) is determined recursively with respect to the 2-adic valuation: if $4 \mid n, then r_3(n) = r_3(n/4); otherwise, r_3(n) = 0 if n \equiv 7 \pmod{8}, r_3(n) = 24 H(n) if n \equiv 3 \pmod{8}, and r_3(n) = 12 H(4n) if n \equiv 1, 2, 5, or $6 \pmod{8}. This formula originates from Gauss's investigations into forms and was refined using the Hurwitz class number in modern treatments. For squarefree n > 4 not of the form $8b + 7, the formula simplifies using ordinary class numbers h(\cdot) of imaginary quadratic fields: r_3(n) = 24 h(-n) if n \equiv 3 \pmod{8}, and r_3(n) = 12 h(-4n) if n \equiv 1, 2, 5, or $6 \pmod{8}, with r_3(n) = 0 if n \equiv 7 \pmod{8}. These expressions connect r_3(n) directly to the arithmetic of quadratic fields, highlighting the deep interplay between sums of squares and . Examples illustrate the formula's application. For n=1 \equiv 1 \pmod{8}, r_3(1) = 12 H(4) = 12 \times \frac{1}{2} = 6, corresponding to the six permutations of (\pm 1, 0, 0). For n=3 \equiv 3 \pmod{8}, r_3(3) = 24 H(3) = 24 \times \frac{1}{3} = 8, from the eight sign combinations of (1, 1, 1). For n=4 = 4 \times 1, r_3(4) = r_3(1) = 6, from permutations of (\pm 2, 0, 0). For n=7 \equiv 7 \pmod{8}, r_3(7) = 0, consistent with the .

For k=4

The sum of squares function for k=4, denoted r_4(n), counts the number of integer solutions (x_1, x_2, x_3, x_4) \in \mathbb{Z}^4 to the equation x_1^2 + x_2^2 + x_3^2 + x_4^2 = n. Jacobi's four-square theorem gives an explicit formula for r_4(n) in terms of the divisors of n. Specifically, r_4(n) = 8 \sum_{\substack{d \mid n \\ 4 \nmid d}} d, where the sum runs over all positive divisors d of n that are not divisible by 4. This general formula admits equivalent expressions depending on the parity of n. When n is , all divisors are odd and thus not divisible by 4, so r_4(n) = 8 \sigma(n), where \sigma(n) denotes the sum of the positive divisors of n. When n is even, write n = 2^r m with m odd and r \geq 1; then r_4(n) = 24 \sum_{\substack{d \mid n \\ d \ odd}} d = 24 \sigma(m), where the sum is over the odd positive divisors of n. These forms are equivalent to the general expression, as the condition $4 \nmid d excludes higher powers of 2 in the even case while weighting the odd divisors appropriately. The theorem refines by not only confirming that r_4(n) > 0 for all positive integers n (since \sigma(n) \geq n + 1 > 0), but also quantifying the exact multiplicity of representations. For instance, r_4(1) = 8, corresponding to the eight permutations of (\pm 1, 0, 0, 0). For n=2, r_4(2) = 24, arising from the 24 signed permutations of (\pm 1, \pm 1, 0, 0). Proofs of the theorem typically employ the \vartheta(z) = \sum_{k=-\infty}^{\infty} e^{2\pi i k^2 z} and its , whose Fourier coefficients yield r_4(n), or alternatively use identities like the .

For k=6

The explicit formula for r_6(n), the number of integer solutions to x_1^2 + x_2^2 + x_3^2 + x_4^2 + x_5^2 + x_6^2 = n, was established by Jacobi in 1829. It takes the form r_6(n) = 16 \sum_{d \mid n} \chi\left( \frac{n}{d} \right) d^2 - 4 \sum_{d \mid n} \chi(d) d^2, where the sums run over the positive divisors d of n, and \chi is the non-principal Dirichlet character modulo 4, given by \chi(m) = \begin{cases} 0 & \text{if } m \equiv 0 \pmod{2}, \\ 1 & \text{if } m \equiv 1 \pmod{4}, \\ -1 & \text{if } m \equiv 3 \pmod{4}. \end{cases} The formula arises from the coefficient extraction in the expansion of the of the Jacobi theta function \vartheta_3(q) = \sum_{m=-\infty}^{\infty} q^{m^2}, equated to a of weighted sums via identities. Analytic proofs rely on the convergence of theta series and Poisson summation, while elementary proofs use bijections between representations and counts, avoiding . Since r_6(n) is a , its value for general n = 2^\alpha \prod p_i^{\beta_i} q_j^{\gamma_j} (with primes p_i \equiv 1 \pmod{4}, q_j \equiv 3 \pmod{4}) factors into local components computable from the applied to each . For instance, r_6(1) = 12, accounting for the six choices of position for \pm 1 and zeros elsewhere; r_6(5) = 312, from the applied to divisors 1 and 5 (both \equiv 1 \pmod{4}), corresponding to representations such as permutations of (\pm 2, \pm 1, 0, 0, 0, 0) and (\pm 1, \pm 1, \pm 1, \pm 1, \pm 1, 0). These values illustrate how the \chi encodes to distinguish residue classes modulo 4. Jacobi's result extends by providing an exact count rather than mere existence, and it has influenced subsequent work on higher even k, though formulas grow more intricate beyond k=8. An accessible proof, emphasizing divisor identities without theta functions, appears in the work of Alaca, Alaca, and Williams (2007).

For k=8

The explicit for r_8(n), the number of integer solutions to x_1^2 + x_2^2 + \dots + x_8^2 = n counting orders and signs, was discovered by Jacobi in 1829 as part of his work on theta functions and elliptic integrals. This expresses r_8(n) directly in terms of the divisors of n, highlighting the nature of representations by eight squares. The formula is given by r_8(n) = 16 \sum_{d \mid n} (-1)^{n + d} d^3, where the sum runs over all positive divisors d of n. This identity relies on the generating function \theta_3(q)^8, where \theta_3(q) = \sum_{m=-\infty}^{\infty} q^{m^2} is the Jacobi theta function, and employs modular form techniques to extract the coefficient of q^n. For odd n, all divisors d are odd, making n + d even and (-1)^{n + d} = 1, so the formula simplifies to r_8(n) = 16 \sigma_3(n), with \sigma_3(n) = \sum_{d \mid n} d^3 denoting the sum of cubes of divisors. For example, when n = 1, the sole divisor is 1, yielding \sigma_3(1) = 1 and r_8(1) = 16, corresponding to the 16 choices of position and sign for a single \pm 1 with seven zeros. For even n, the alternating sign introduces cancellations based on the parities of divisors, but the overall value remains positive, ensuring every positive integer can be expressed as a sum of eight squares. Jacobi's derivation connects to the theory of forms and has implications for the of sums of squares, as facilitated by Degen's eight-square , which states that the product of two sums of eight squares is again a sum of eight squares. This underpins the multiplicative structure observed in the for r_8(n). The result extends Jacobi's earlier formulas for k = 2, 4, 6, completing the set of explicit expressions for small even k.

Numerical Values and Computation

Tables of Small Values

The sum of squares function r_k(n) gives the number of solutions to x_1^2 + x_2^2 + \dots + x_k^2 = n, where matters, signs are distinguished, and zeros are allowed. For small values of n, these can be computed using explicit formulas or theta series expansions. The following tables list r_k(n) for n = 0 to $10 and selected even k, as these admit particularly simple closed-form expressions derived from Jacobi's theorems.

For k = 2

The values follow Jacobi's two-square theorem: r_2(n) = 4 (d_1(n) - d_3(n)), where d_i(n) is the number of divisors of n congruent to i \pmod{4}.
nr_2(n)
01
14
24
30
44
58
60
70
84
94
108

For k = 4

The values follow Jacobi's four-square theorem: r_4(n) = 8 \sum_{d \mid n, \, 4 \nmid d} d if n is odd, and r_4(n) = 24 \sum_{d \mid n, \, d \odd} d if n is even.
nr_4(n)
01
18
224
332
424
548
696
764
824
9104
10144

For k = 6

The values follow Jacobi's six-square theorem, involving character sums: r_6(n) = 16 \sum_{d \mid n} \chi(d') d^2 - 4 \sum_{d \mid n} \chi(d) d^2, where \chi is the non-principal modulo 4 and d' = d if d odd, d/4 if d \equiv 0 \pmod{4}.
nr_6(n)
01
112
260
3160
4252
5312
6544
7960
81020
9876
101560

For k = 8

The values follow Jacobi's eight-square theorem: r_8(n) = 16 \sum_{d \mid n} (-1)^{n+d} d^3. This simplifies to $16 \sigma_3(n) for odd n, where \sigma_3(n) = \sum_{d \mid n} d^3, and adjusted forms for even n.
nr_8(n)
01
116
2112
3448
4240
5960
62016
71792
8672
93456
105376

Algorithms for Computation

The sum of squares function r_k(n), which counts the number of integer solutions to x_1^2 + \cdots + x_k^2 = n (allowing zeros, distinguishing order and signs), can be computed exactly using number-theoretic algorithms that leverage explicit formulas for small fixed k. These formulas, derived from the theory of modular forms and theta series, express r_k(n) as a weighted sum over the divisors of n, making computation efficient provided n is factorized. Factorization of n can be performed using standard algorithms such as trial division up to \sqrt{n} or more advanced methods like the quadratic sieve for large n; once the prime factorization is known, all divisors can be generated in O(d(n)) time, where d(n) is the number of divisors, typically small (e.g., d(n) = O(n^\epsilon) for any \epsilon > 0). For even k \leq 8, Jacobi provided closed-form expressions in terms of divisor sums, which form the basis of the most efficient algorithms for these cases. For k=2, Gauss and Jacobi showed r_2(n) = 4(d_1(n) - d_3(n)), where d_i(n) counts the divisors of n congruent to i \pmod{4}. To compute this, list all divisors, classify them modulo 4, and apply the difference; this takes O(d(n)) arithmetic operations after factorization. Similarly, for k=4, Jacobi's formula is r_4(n) = 8 \sum_{d \mid n, \, 4 \nmid d} d if n is odd, and adjusted by subtracting contributions from powers of 4 for even n. For k=6 and k=8, the formulas involve higher powers of divisors weighted by characters modulo 4 or parity: r_6(n) = 16 \sum_{d \mid n} (-1)^{n+d} d^2 - 4 \sum_{d \mid n} (-1)^{(d-1)/2} d^2 (for odd divisors in the second sum), and r_8(n) = 16 \sum_{d \mid n} (-1)^{n+d} d^3, respectively. These require generating divisors and evaluating the sums, again in O(d(n)) time, with the dominant cost being factorization for large n. These methods are highly practical, enabling computation of r_k(n) for n up to $10^{20} or more on modern hardware when paired with efficient factoring libraries. For odd k, such as k=3, no simple exists without additional , but an explicit expression involves the Hurwitz class number: for example, when n is square-free, r_3(n) = 12 h(-4n) if n \not\equiv 7 \pmod{8} (with r_3(n) = 0 if n \equiv 7 \pmod{8} or of form 4^a(8b+7)), or r_3(n) = 24 h(-n) if n \equiv 3 \pmod{8}; more generally, it can be expressed using the Hurwitz class number H(-4n). Computing class numbers requires solving the problem in quadratic fields \mathbb{Q}(\sqrt{-d}), which can be done in subexponential time using algorithms like the method or Buchmann's algorithm, feasible for discriminants up to $10^{12} or so. For n not of the form $4^a(8b+7) (by Legendre's theorem), where r_3(n) > 0, probabilistic methods can also find representations quickly by searching small values and reducing to two-square problems, but counting all requires the class number approach. For general k or when explicit formulas are unavailable (e.g., k > 8), a recursive dynamic programming provides an exact count. Define r_k(m) recursively as r_k(m) = \sum_{j=-\lfloor \sqrt{m} \rfloor}^{\lfloor \sqrt{m} \rfloor} r_{k-1}(m - j^2), with base case r_1(m) = 2 if m is a positive , $1 if m=0, and $0 otherwise. This can be implemented by computing a table of all r_l(t) for $1 \leq l \leq k and $0 \leq t \leq n, updating via with squares: for each l, iterate over possible j^2 \leq t and add to the table. The is O(k n^{3/2}) in the naive form, but optimizations like precomputing squares reduce constants; for fixed small k and large n, this is slower than methods but scales well for moderate n (e.g., n \leq 10^6) and arbitrary k. For very large n, hybrid approaches combine asymptotics r_k(n) \sim \frac{\pi^{k/2}}{\Gamma(k/2)} n^{k/2 - 1} with exact corrections via modular forms, though exact computation remains - or recursion-based.

References

  1. [1]
    [PDF] SUM OF TWO SQUARES Contents 1. Introduction 1 2. Preliminaries ...
    Aug 22, 2008 · I will investigate which numbers can be written as the sum of two squares and in how many ways, providing enough basic number theory so even.
  2. [2]
    [PDF] Sums of Squares - Boston University
    It says that if two numbers are sum of two squares, so is their product. For example 493 = 17 ∗ 29 and 17 = 12 + 42, 29 = 22 + 52 give 493 = (1 ∗ 2 ...
  3. [3]
    Sum of Squares Function -- from Wolfram MathWorld
    The number of representations of n by k squares, allowing zeros and distinguishing signs and order, is denoted r_k(n). The special case k=2 corresponding to ...
  4. [4]
    New Infinite Families of Exact Sums of Squares Formulas, Jacobi ...
    explicit exact formulas that generalize Jacobi's (1829) 4 and. 8 squares ... The problem of representing an integer as a sum of squares of integers has ...
  5. [5]
    [PDF] an introduction - theory of numbers
    We shall often use the word 'number' as meaning 'integer' (or 'positive integer', etc.), when it is clear from the context that we are considering only numbers ...
  6. [6]
    [PDF] Introduction to the Theory of Numbers
    Nov 21, 2014 · Our first aim has been to Write an interesting book, and one unlike other books. We may have succeeded at the price of too much eccen- tricity, ...
  7. [7]
    [PDF] sums of squares and partition congruences
    For positive integers n and k, let rk(n) denote the number of represen- tations of n as a sum of k squares, where representations with different orders and.
  8. [8]
    Jacobi Theta Functions -- from Wolfram MathWorld
    The Jacobi theta functions are the elliptic analogs of the exponential function, and may be used to express the Jacobi elliptic functions.
  9. [9]
    [PDF] THETA FUNCTIONS 1. First Example We have seen one way to ...
    Let rk(n) be the number of ways to represent n as a sum of k squares of integers. Then we see directly that. X n≥0 rk(n)qn = ϑk(τ). Sums of 1 square are ...
  10. [10]
    [PDF] Modular Forms and Jacobi's Four Square Theorem - Princeton Math
    We would like to have a generating function for the number of ways to write a number as a sum of k squares. One particular example of a theta function does the ...
  11. [11]
    [PDF] 10 Applications of Theta - Functions - Princeton University
    A more refined analysis requires the transformation properties of the generating function which goes back ... sum of two squares, that is, c = m2 + n2 with ...
  12. [12]
    [PDF] Modular Forms and Sums of Squares - Berkeley Math
    We say that a modular form is a cusp form if a0 = 0, or equivalently if limIm(z)→∞ f(z) = 0. We denote by Sk(SL2(Z)) the graded ideal of weight-k cusp forms. I.
  13. [13]
    [PDF] The Jacobi ϑ Function - Simon Rubinstein-Salzedo
    Jun 13, 2005 · Jacobi ϑ function very useful for solving problems dealing with sums of squares. To this effect, define a representation function rh : N ...<|control11|><|separator|>
  14. [14]
    [PDF] SUM OF TWO SQUARES Contents 1. Meromorphic continuation of ...
    It turns out that the Fourier expansion of the square of Jacobi theta function is related to the following classical problem in number theory:
  15. [15]
    [PDF] A simple proof of Jacobi's four-square theorem
    Recently [2], I showed how one can obtain Jacobi's two-square theorem from the triple-product identity. In this note I show that the triple-product identity.
  16. [16]
    [PDF] Jacobi's Four Squares Theorem - Digital Commons at Oberlin
    An integer n > 1 can be written as a sum of two squares if and only if the prime factorization of n does not contain pk (where k is the multiplicity of p) for.
  17. [17]
    [PDF] The Parents of Jacobi's Four Squares Theorem Are Unique
    Jacobi's four squares theorem asserts that the number of representations of a posi- tive integer n as a sum of four squares is 8 times the sum of the positive ...Missing: Carl 1834
  18. [18]
    [PDF] the simplest proof of jacobi's six squares theorem
    Jun 1, 2007 · A very simple proof is given of Jacobi's formula for the number of representations of a positive integer as the sum of six squares. 1.Missing: r_6( | Show results with:r_6(
  19. [19]
    Jacobi's Four and Eight Squares Theorems and Partitions into ...
    Feb 7, 2019 · formulas for r4(n) and r8(n): r4(n)=8. ∑. 4\d|n d, and r8(n) = 16. ∑ d|n. (−1)n+dd3. The formula for r4(n) can be expressed in the following ...Missing: r_8( | Show results with:r_8(
  20. [20]
    [PDF] An addition formula for the Jacobian theta function - arXiv
    (Jacobi's Eight Squares Theorem) For each positive integer n, we have r8(n) = 16 ∑ d|n. (-1)n+dd3. Proof. Applying the infinite product representations of ...
  21. [21]
    A004018 - OEIS
    Hardy and E. M. Wright, An Introduction to the Theory of Numbers. 3rd ed., Oxford Univ. Press, 1954, p. 240, r(n). W. König and J. Sprekels, Karl Weierstraß ...
  22. [22]
    A000118 - OEIS
    Spearman and Kenneth S. Williams, The simplest arithmetic proof of Jacobi's four squares theorem, Far East J. of Math. Sci. 2.3 (2000): 433-440.<|separator|>
  23. [23]
    A000141 - OEIS
    Number of ways of writing n as a sum of 6 squares. 20. 1, 12, 60, 160, 252, 312, 544, 960, 1020, 876, 1560, 2400, 2080, 2040, 3264, 4160, 4092, 3480, 4380 ...Missing: theta | Show results with:theta
  24. [24]