Fact-checked by Grok 2 weeks ago

Modular group

The modular group, often denoted \mathrm{PSL}(2, \mathbb{Z}), is the projective special linear group consisting of all $2 \times 2 matrices with integer entries and determinant 1, taken modulo the center \{\pm I\}.^[1] Equivalently, it is the quotient \mathrm{SL}(2, \mathbb{Z}) / \{\pm I\}, where \mathrm{SL}(2, \mathbb{Z}) is the special linear group of $2 \times 2 integer matrices of determinant 1 under matrix multiplication.^[2] This group serves as a discrete subgroup of \mathrm{PSL}(2, \mathbb{R}) and is fundamental in connecting number theory, geometry, and analysis through its actions and representations.^[3] The modular group admits a concrete presentation via generators S = \begin{pmatrix} 0 & -1 \\ 1 & 0 \end{pmatrix} (of order 4 in \mathrm{SL}(2, \mathbb{Z}), corresponding to order 2 in \mathrm{PSL}(2, \mathbb{Z})) and T = \begin{pmatrix} 1 & 1 \\ 0 & 1 \end{pmatrix} (of infinite order), satisfying the relations S^4 = I, (ST)^3 = S^2 in \mathrm{SL}(2, \mathbb{Z}).^[3] Abstractly, \mathrm{PSL}(2, \mathbb{Z}) is isomorphic to the free product \mathbb{Z}/2\mathbb{Z} * \mathbb{Z}/3\mathbb{Z}, highlighting its structure as a non-abelian group generated by elements of orders 2 and 3.^[1] Elements of the group are classified by their trace: elliptic if |\mathrm{tr}(\gamma)| < 2, parabolic if |\mathrm{tr}(\gamma)| = 2, and hyperbolic if |\mathrm{tr}(\gamma)| > 2.^[1] The modular group acts on the upper half-plane \mathbb{H} = \{ z = x + iy \mid y > 0 \} via Möbius transformations z \mapsto \frac{az + b}{cz + d} for \gamma = \begin{pmatrix} a & b \\ c & d \end{pmatrix} \in \mathrm{SL}(2, \mathbb{Z}), preserving the hyperbolic metric and yielding a discontinuous action whose quotient is the modular surface, a Riemann surface of genus 0.^[2] This action underpins the theory of modular forms—holomorphic functions on \mathbb{H} invariant under the group—and links to elliptic curves, the Langlands program, and classifications of complex tori.^[2] Historically, the group traces conceptual roots to Euclid's algorithm via continued fractions and was rigorously developed in the late 19th century by Felix Klein and Henri Poincaré as a prototype for discrete group actions in geometry.^[1]

Definition

Matrix Presentation

The special linear group \mathrm{SL}(2, \mathbb{Z}) consists of all $2 \times 2 matrices with integer entries and determinant $1$.^[4] These matrices form a group under matrix multiplication.^[5] The modular group, denoted \mathrm{PSL}(2, \mathbb{Z}), is obtained as the quotient group \mathrm{SL}(2, \mathbb{Z}) / \{\pm I\}, where I is the $2 \times 2 identity matrix and \{\pm I\} is the center of \mathrm{SL}(2, \mathbb{Z}).^[4] This quotient identifies each matrix A \in \mathrm{SL}(2, \mathbb{Z}) with -A, ensuring that elements of \mathrm{PSL}(2, \mathbb{Z}) correspond to equivalence classes of such matrices. Explicit examples include the equivalence class of the identity matrix

\begin{pmatrix} 1 & 0 \\ 0 & 1 \end{pmatrix},

which represents the identity element, and the class of the inversion matrix

\begin{pmatrix} 0 & -1 \\ 1 & 0 \end{pmatrix},

which has order $2$ in the group.^[5] The group \mathrm{PSL}(2, \mathbb{Z}) acts on the upper half-plane via Möbius transformations: for a representative matrix \begin{pmatrix} a & b \\ c & d \end{pmatrix} \in \mathrm{SL}(2, \mathbb{Z}), the action on z \in \mathbb{H} is given by

z \mapsto \frac{az + b}{cz + d}.

^[4] This action is well-defined on the quotient since \pm I act trivially.^[5]

Generators and Elements

The modular group \mathrm{PSL}(2, \mathbb{Z}) is generated by two elements S and T, where S is the image of the matrix \begin{pmatrix} 0 & -1 \\ 1 & 0 \end{pmatrix} (of order 2) and T is the image of \begin{pmatrix} 1 & 1 \\ 0 & 1 \end{pmatrix} (of infinite order). These generators correspond to the linear fractional transformations S(z) = -1/z and T(z) = z + 1, respectively. Every element of the modular group can be uniquely expressed as a finite word in S and T, up to the relations in the group presentation. Elements of the modular group are classified according to the trace of their lifts to matrices in \mathrm{SL}(2, \mathbb{Z}): parabolic if |\mathrm{tr}| = 2, elliptic if |\mathrm{tr}| < 2 (specifically |\mathrm{tr}| = 0 or $1, since traces are integers), and hyperbolic if |\mathrm{tr}| > 2$. Parabolic elements fix exactly one point on the boundary of the upper half-plane (a cusp), elliptic elements fix points inside the upper half-plane (with finite order), and hyperbolic elements fix two points on the boundary. The word problem in the modular group—determining whether two words in S and T represent the same element—can be solved algorithmically using continued fraction expansions of rational numbers or properties of cusp widths in the fundamental domain. Specifically, each matrix in \mathrm{SL}(2, \mathbb{Z}) maps \infty to a rational number whose continued fraction expansion encodes the sequence of generators S and T via paths in the Farey tesselation or Stern–Brocot tree. Representative examples include translations, which are powers T^n = \begin{pmatrix} 1 & n \\ 0 & 1 \end{pmatrix} for n \in \mathbb{Z} (parabolic elements shifting cusps by integers); the inversion S (elliptic of order 2); and products such as ST = \begin{pmatrix} 0 & -1 \\ 1 & 1 \end{pmatrix} (elliptic of order 3, representing a rotation by $120^\circ). Hyperbolic elements arise as longer products, for example \begin{pmatrix} 2 & 1 \\ 1 & 1 \end{pmatrix} with trace $3 > 2, corresponding to expansions along geodesics.^[5]

Algebraic Properties

Group Presentation

The modular group \mathrm{PSL}(2, \mathbb{Z}) has the abstract presentation \langle S, T \mid S^2 = (ST)^3 = 1 \rangle, where S and T are generators satisfying these relations and no others. This presentation captures the structure of the group as the projective special linear group over the integers, modulo the center \{\pm I\}. The relations derive directly from the standard matrix generators: S = \begin{pmatrix} 0 & -1 \\ 1 & 0 \end{pmatrix} and T = \begin{pmatrix} 1 & 1 \\ 0 & 1 \end{pmatrix}. Direct computation yields S^2 = -I, which represents the identity in \mathrm{PSL}(2, \mathbb{Z}) since scalar multiples by -1 are quotiented out; similarly, (ST)^3 = -I \equiv I in the projective group. These matrix multiplications confirm that S has order 2 and ST has order 3 in \mathrm{PSL}(2, \mathbb{Z}). This presentation establishes an isomorphism \mathrm{PSL}(2, \mathbb{Z}) \cong \mathbb{Z}/2\mathbb{Z} * \mathbb{Z}/3\mathbb{Z}, the free product of the cyclic groups of orders 2 and 3, amalgamated over their trivial subgroup. To see this, map the free product to \mathrm{PSL}(2, \mathbb{Z}) by sending the generator \alpha of \mathbb{Z}/2\mathbb{Z} to the class of S and the generator \beta of \mathbb{Z}/3\mathbb{Z} to the class of ST; this homomorphism is surjective since S and T generate the group. Injectivity follows from the unique representation of elements as reduced words alternating between powers of \alpha (order 2) and powers of \beta (order 3), with no nontrivial relations beyond those given; any such word mapping to the identity would contradict the faithful action on the real line, where nontrivial words shift intervals like (0, \infty) or (-\infty, 0) nontrivially. Thus, the group is freely generated by these cyclic factors modulo the specified relations.

Quotients and Subgroups

The principal congruence subgroups of the modular group PSL(2, ℤ) are defined as the kernels of the natural surjective homomorphisms PSL(2, ℤ) → PSL(2, ℤ/Nℤ) for each positive integer N, where the map is induced by reduction modulo N. These subgroups, denoted Γ(N), are normal in PSL(2, ℤ) since they are kernels of homomorphisms. These homomorphisms yield finite quotients of the modular group isomorphic to PSL(2, ℤ/Nℤ). For example, the quotient PSL(2, ℤ)/Γ(2) is isomorphic to the symmetric group S₃ of order 6. Similarly, PSL(2, ℤ)/Γ(3) is isomorphic to the alternating group A₄ of order 12. The index of Γ(N) in PSL(2, ℤ) is |PSL(2, ℤ/Nℤ)|, which equals N³ ∏{p|N} (1 - 1/p²) for N=1,2 and (1/2) N³ ∏{p|N} (1 - 1/p²) for N ≥ 3. This follows from the index [SL(2, ℤ) : Γ(N)] = N³ ∏_{p|N} (1 - 1/p²) adjusted for the center {±I}, as -I ∉ Γ(N) for N ≥ 3 in SL(2, ℤ), with the product accounting for the structure over prime powers dividing N. Among the congruence subgroups of PSL(2, ℤ), the principal congruence subgroups Γ(N) are the only normal ones; any normal congruence subgroup must be principal. Other congruence subgroups, such as the Hecke subgroups Γ₀(N) and Γ₁(N), exist but are not normal in general and are treated in detail in the section on congruence subgroups. Non-congruence subgroups of finite index also exist, providing further quotients, though they lie outside the scope of congruence properties.

Connections to Other Groups

The modular group \mathrm{PSL}(2, \mathbb{Z}) is a quotient of the braid group B_3 on three strands by its center. The braid group B_3 has presentation \langle \sigma_1, \sigma_2 \mid \sigma_1 \sigma_2 \sigma_1 = \sigma_2 \sigma_1 \sigma_2 \rangle, and its center is infinite cyclic, generated by (\sigma_1 \sigma_2)^3. The quotient map B_3 \to \mathrm{PSL}(2, \mathbb{Z}) sends the center to the identity, yielding an isomorphism B_3 / Z(B_3) \cong \mathrm{PSL}(2, \mathbb{Z}). This relation highlights the modular group as arising from topological structures in low dimensions, with the kernel providing a central cyclic extension. This quotient is in fact the universal central extension of \mathrm{PSL}(2, \mathbb{Z}), meaning B_3 captures all central extensions of the modular group up to isomorphism. The second homology group H_2(\mathrm{PSL}(2, \mathbb{Z}), \mathbb{Z}) is infinite cyclic, classifying such extensions, and B_3 realizes the generator. Seminal work on this connection traces to studies of braid groups and their linear representations, where the modular group emerges as the image under faithful maps preserving the braid relation. The Artin representation provides an explicit embedding of the modular group's structure into the braid group context. The generators of \mathrm{PSL}(2, \mathbb{Z}), typically taken as the order-2 element S (corresponding to the matrix \begin{pmatrix} 0 & -1 \\ 1 & 0 \end{pmatrix} modulo \pm I) and the element U = ST of order 3, lift to braids via the inverse of the quotient map. Specifically, the standard symplectic representation \rho: B_3 \to \mathrm{SL}(2, \mathbb{Z}) sends \sigma_1 \mapsto \begin{pmatrix} 1 & 1 \\ 0 & 1 \end{pmatrix} and \sigma_2 \mapsto \begin{pmatrix} 1 & 0 \\ -1 & 1 \end{pmatrix}, satisfying the braid relation and generating \mathrm{SL}(2, \mathbb{Z}) with kernel the center; projecting to \mathrm{PSL}(2, \mathbb{Z}) gives the desired isomorphism. This representation, rooted in Artin's foundational work on braids and their automorphisms of free groups, underscores the algebraic interplay between topological and linear structures. As a discrete subgroup, \mathrm{PSL}(2, \mathbb{Z}) embeds naturally into the Lie group \mathrm{PSL}(2, \mathbb{R}) via the inclusion \mathbb{Z} \hookrightarrow \mathbb{R}, identifying integer matrices with real ones. This embedding preserves the group operation and makes \mathrm{PSL}(2, \mathbb{Z}) a lattice in \mathrm{PSL}(2, \mathbb{R}), fundamental for its action on the hyperbolic plane. The image consists of all projective transformations with integer coefficients and determinant 1, modulo scalars. The modular group also admits a surjective homomorphism onto the symmetric group S_3. This arises from reduction modulo 2: the map \mathrm{SL}(2, \mathbb{Z}) \to \mathrm{SL}(2, \mathbb{Z}/2\mathbb{Z}) \cong S_3, with kernel the principal congruence subgroup \Gamma(2) of level 2, which has index 6 in \mathrm{SL}(2, \mathbb{Z}); projecting yields \mathrm{PSL}(2, \mathbb{Z}) \to \mathrm{PSL}(2, \mathbb{Z}/2\mathbb{Z}) \cong S_3. This quotient reflects the modular group's richness, as S_3 captures its finite symmetries in characteristic 2.

Arithmetic Properties

Number-Theoretic Interpretations

The modular group \mathrm{SL}(2, \mathbb{Z}) acts on the set of rational numbers \mathbb{Q} \cup \{\infty\} via Möbius transformations \gamma \cdot z = \frac{az + b}{cz + d} for \gamma = \begin{pmatrix} a & b \\ c & d \end{pmatrix} \in \mathrm{SL}(2, \mathbb{Z}) and z \in \mathbb{Q} \cup \{\infty\}. This action preserves the rationals and connects directly to continued fraction expansions. Specifically, the generator T = \begin{pmatrix} 1 & 1 \\ 0 & 1 \end{pmatrix} corresponds to adding 1 to the first partial quotient of the continued fraction of z, effectively shifting the expansion, while the generator S = \begin{pmatrix} 0 & -1 \\ 1 & 0 \end{pmatrix} takes the reciprocal, which inverts the continued fraction by transforming [a_0; a_1, a_2, \dots] to [0; a_0, a_1, \dots] or similar adjustments depending on the sign. These operations generate all continued fraction expansions of rationals through compositions, linking the group structure to the Stern-Brocot tree and Farey sequences, where adjacent fractions differ by the action of ST or similar elements.^[3]^[6] Hyperbolic elements in \mathrm{SL}(2, \mathbb{Z}), characterized by trace greater than 2 in absolute value, play a key role in classifying quadratic irrationals. The fixed points of such an element \gamma solve \gamma \cdot z = z, yielding the quadratic equation cz^2 + (d - a)z - b = 0 with integer coefficients and discriminant \mathrm{tr}(\gamma)^2 - 4 > 0, not a perfect square. Thus, these fixed points are precisely the real quadratic irrationals, and every quadratic irrational arises as a fixed point of some hyperbolic element, often paired with its Galois conjugate as the two real fixed points of \gamma. This classification underscores the arithmetic nature of hyperbolic orbits, distinguishing them from parabolic (rational fixed points) and elliptic (complex fixed points) elements.^[7]^[8] The modular group \mathrm{SL}(2, \mathbb{Z}) also acts on the space of binary quadratic forms Q(x, y) = ax^2 + bxy + cy^2 with integer coefficients by substitution: for \gamma = \begin{pmatrix} \alpha & \beta \\ \gamma & \delta \end{pmatrix}, the transformed form is Q((x, y) \gamma) = a(\alpha x + \beta y)^2 + b(\alpha x + \beta y)(\gamma x + \delta y) + c(\gamma x + \delta y)^2, preserving the discriminant d = b^2 - 4ac. Two forms are equivalent if one arises from the other via this action, partitioning the forms into equivalence classes that represent the same set of integer values. For positive definite forms (d < 0), a form is reduced if it satisfies |b| \leq a \leq c and b \geq 0 when a = c or a = |b|; every such form is equivalent under \mathrm{SL}(2, \mathbb{Z}) to a unique reduced form.^[9]^[10] The reduced forms serve as canonical representatives for the equivalence classes, and their enumeration for a fixed discriminant d yields the class number h(d), the number of distinct classes. This counting is facilitated by the fundamental domain of the \mathrm{SL}(2, \mathbb{Z})-action on the associated upper half-plane, where reduced forms correspond to points in this domain via the map Q \mapsto \tau_Q = \frac{-b + \sqrt{d}}{2a}, ensuring no two reduced forms are equivalent and covering all classes exhaustively. This structure, originating in Gauss's work, links the class number to the arithmetic of quadratic fields, with h(d) finite and computable via reduction.^[9]^[10]

Congruence Subgroups

Congruence subgroups of the modular group PSL(2, ℤ) are finite-index subgroups that contain the principal congruence subgroup Γ(N) for some positive integer N, known as the level of the subgroup. These subgroups arise as kernels of natural reduction modulo N homomorphisms from PSL(2, ℤ) to PSL(2, ℤ/Nℤ), providing an arithmetic structure to the modular group.^[11] The principal congruence subgroup of level N is defined as Γ(N) = ker(PSL(2, ℤ) → PSL(2, ℤ/Nℤ)), consisting of elements congruent to the identity matrix modulo N. Other standard congruence subgroups include Γ₀(N), the subgroup of matrices \begin{pmatrix} a & b \\ c & d \end{pmatrix} in SL(2, ℤ) with c ≡ 0 mod N (projected to PSL(2, ℤ)); Γ₁(N), where additionally a ≡ d ≡ 1 mod N; and the general form Γ_H(N), where H is a subgroup of (ℤ/Nℤ)× and the matrices satisfy a ≡ d mod N with a - 1 ∈ H, and c ≡ 0 mod N. These form a nested chain Γ(N) ⊂ Γ₁(N) ⊂ Γ₀(N) ⊂ SL(2, ℤ) for N ≥ 1. The index [PSL(2, ℤ) : Γ₀(N)] equals N ∏_{p ∣ N} (1 + 1/p), where the product is over distinct primes dividing N; this coincides with the index in SL(2, ℤ) since the center {±I} lies in both groups.^[12] The geometry of these subgroups is reflected in the modular curves they define, such as X₀(N) = Γ₀(N) \ ℍ^, where ℍ^ is the extended upper half-plane. The number of cusps of Γ₀(N), which equals the number of Γ₀(N)-orbits on ℚ ∪ {∞}, is given by ∑_{d ∣ N} φ(gcd(d, N/d)), with φ the Euler totient function.^[13] The genus of the modular curve X(Γ) for a congruence subgroup Γ of level N is computed via the Riemann-Hurwitz formula applied to the projection from X(Γ) to X(1), yielding g = 1 + μ/12 - ν₂/4 - ν₃/3 - ν_∞/2, where μ = [PSL(2, ℤ) : Γ] is the index, ν₂ and ν₃ are the numbers of elliptic points of orders 2 and 3, respectively, and ν_∞ is the number of cusps (often denoted c, with ε sometimes summarizing the elliptic contributions as ε/4 ≈ ν₂/4 + ν₃/3).^[14] For Γ₀(N), explicit formulas exist for ν₂ = ∏{p ∣ N} (1 + (-1/p)) and ν₃ = ∏{p ∣ N} (1 + (-3/p)), where (·/p) is the Legendre symbol.^[14] All congruence subgroups are finitely generated, as they have finite index in the finitely generated group PSL(2, ℤ), and their abelianizations are known and computable from the Reidemeister-Schreier method applied to the group's presentation.^[15] For example, the abelianization of Γ(N) for N ≥ 3 is (ℤ/Nℤ)^2.^[11]

Geometric Properties

Action on the Upper Half-Plane

The modular group \Gamma = \mathrm{PSL}(2, \mathbb{Z}) acts faithfully on the upper half-plane \mathbb{H} = \{ z \in \mathbb{C} \mid \operatorname{Im} z > 0 \} by Möbius transformations: for \gamma = \begin{pmatrix} a & b \\ c & d \end{pmatrix} \in \Gamma, the action is defined by \gamma \cdot z = \frac{az + b}{cz + d}, where a, b, c, d \in \mathbb{Z}, ad - bc = 1, and matrices are identified modulo \{\pm I\}.^[16] This action preserves the hyperbolic metric on \mathbb{H}, given infinitesimally by ds^2 = \frac{dx^2 + dy^2}{y^2} for z = x + iy, ensuring that \mathbb{H} serves as a model for the hyperbolic plane.^[16] The corresponding hyperbolic distance between points z, w \in \mathbb{H} is

d(z, w) = \arcosh\left(1 + \frac{|z - w|^2}{2 \operatorname{Im} z \cdot \operatorname{Im} w}\right),

which is invariant under the group action and quantifies the geometry induced on orbits.^[17] Elements of \Gamma are classified by their fixed points under this action: elliptic elements fix a point in \mathbb{H}, parabolic elements fix exactly one point on the boundary \mathbb{R} \cup \{\infty\} (known as cusps), and hyperbolic elements fix two points on the boundary.^[18] The elliptic fixed points include i, stabilized by the order-2 cyclic subgroup generated by the transformation z \mapsto -1/z, and \rho = e^{2\pi i / 3} = -\frac{1}{2} + i \frac{\sqrt{3}}{2}, stabilized by the order-3 cyclic subgroup generated by z \mapsto -\frac{1}{z+1}.^[16] Parabolic fixed points occur at the cusps, such as \infty under translations z \mapsto z + n for n \in \mathbb{Z}, while hyperbolic elements have no fixed points in \mathbb{H} but pair boundary points.^[18] By the orbit-stabilizer theorem, the stabilizer of an elliptic point in \Gamma is a finite cyclic group, whose order equals the order of the elliptic element fixing that point, leading to discrete orbits for generic points in \mathbb{H}.^[16] The action is properly discontinuous on \mathbb{H}, meaning that for any compact subset K \subset \mathbb{H}, only finitely many group elements map K to intersect itself, which ensures that orbits are discrete and allows the construction of fundamental domains where the action tiles \mathbb{H} without interior overlaps.^[16] This discontinuity property underpins the geometric realization of \Gamma as a Fuchsian group of the first kind.^[16]

Hyperbolic Tessellations

The modular group \mathrm{PSL}(2,\mathbb{Z}) acts on the hyperbolic plane \mathbb{H}, identified with the upper half-plane, via Möbius transformations, inducing a tessellation that covers \mathbb{H} without interior overlaps. The standard fundamental domain for this action is the region D = \{ z \in \mathbb{H} \mid |\operatorname{Re} z| \leq 1/2, |z| \geq 1 \}. This curvilinear triangle has vertices at the elliptic fixed points \rho = e^{2\pi i / 3} and i, extending to the cusp at infinity along the vertical boundaries.^[19] The boundary of D consists of two infinite vertical geodesics at \operatorname{Re} z = \pm 1/2 (for |z| \geq 1) and the circular arc |z| = 1 connecting \rho to \rho + 1. These boundaries are identified via group elements: the vertical geodesics are paired by the parabolic generator T: z \mapsto z + 1, while the arc segments are paired by the elliptic generator S: z \mapsto -1/z (mapping the left half-arc to the right) and the composition ST: z \mapsto -1/(z + 1). These identifications ensure that adjacent copies of D glue seamlessly along their boundaries, reflecting the group's presentation.^[19] The full tessellation arises as the disjoint union \mathbb{H} = \bigcup_{\gamma \in \mathrm{PSL}(2,\mathbb{Z})} \gamma D, where each \gamma D is a copy of the fundamental domain. This partitions \mathbb{H} into ideal triangular tiles, each with vertices at rational points on the real line (corresponding to cusps) and angles of \pi/2 or \pi/3 at elliptic points; the fundamental domain itself decomposes into two such congruent ideal triangles joined along the geodesic from i to the arc's midpoint. This Farey tessellation visualizes the group's orbits and underlies many geometric properties of modular forms.^[20] The quotient orbifold \mathbb{H} / \mathrm{PSL}(2,\mathbb{Z}), known as the modular surface, inherits singularities from the action: an elliptic point of order 2 at the image of i, an elliptic point of order 3 at the image of \rho, and a single cusp corresponding to the orbit of infinity. By the Gauss-Bonnet theorem for orbifolds, the hyperbolic area of this surface is \pi/3, computed as $2\pi times the orbifold Euler characteristic $1/6 (accounting for the sphere topology adjusted by the puncture and elliptic orders). This finite area confirms the cofiniteness of the action and quantifies the "volume" of the moduli space of elliptic curves.^[21]

Generalizations

Hecke Groups

The Hecke groups provide a natural generalization of the modular group to a broader class of discrete subgroups of the projective special linear group PSL(2, ℝ), acting on the upper half-plane via Möbius transformations. Introduced by Erich Hecke in his study of Dirichlet series satisfying functional equations, these groups are defined for integer parameters q ≥ 3 as H_q = ⟨S, U_λ | S² = (S U_λ)^q = 1⟩, where S(z) = -1/z and U_λ(z) = z + λ with λ = 2 cos(π/q).^[22] This presentation mirrors the structure of the modular group's presentation ⟨S, T | S² = (S T)^3 = 1⟩, where T(z) = z + 1 corresponds to the case q = 3 and λ = 1, yielding H_3 ≅ PSL(2, ℤ). For integer q ≥ 3, the Hecke group H_q is isomorphic to the free product ℤ₂ * ℤ_q, reflecting its geometric realization as the orientation-preserving index-2 subgroup of the (2, q, ∞) triangle group in hyperbolic geometry. Examples include q = 4 with λ = √2, yielding a group related to square tessellations, and q = 6 with λ = √3, connected to hexagonal structures. These isomorphisms highlight the combinatorial freedom in the group structure, where the ℤ_q factor arises from the rotational symmetry of order q around a fixed point of S U_λ. The definition extends to real q > 2 by setting λ = 2 cos(π/q), producing discrete groups for 1 ≤ λ < 2, or more generally for any λ ≥ 2, where the generators S and U_λ yield a Fuchsian group of the first kind without the finite-order relation on S U_λ. Hecke established that such groups are discrete in PSL(2, ℝ) if and only if λ ≥ 2 or λ = 2 cos(π/q) for integer q ≥ 3.^[23] In the limiting case q → ∞, λ → 2, and H_∞ is the free product ℤ₂ * ℤ generated by S and U_2(z) = z + 2 with presentation ⟨S, U_2 | S² = 1⟩.^[23] The fundamental domains for H_q resemble that of the modular group but feature adjusted widths to accommodate the translation length λ. Specifically, a standard fundamental domain is the hyperbolic region {z ∈ ℍ : |Re z| ≤ λ/2, |z| ≥ 1}, where ℍ denotes the upper half-plane, with side pairings induced by S on the arc |z| = 1 and by U_λ on the vertical lines Re z = ±λ/2.^[23] This domain tiles the hyperbolic plane under the H_q-action, with the width λ/2 determining the strip's breadth and influencing properties like cusp widths and tessellation patterns. For λ = 2 in H_∞, the domain simplifies to a strip of width 1, emphasizing the infinite nature of the rotational component.^[23]

Dyadic Monoid

The dyadic monoid is the subsemigroup of the modular group consisting of all matrices in \mathrm{SL}(2, \mathbb{Z}) with non-negative integer entries and determinant 1.^[24] These matrices represent positive orientation-preserving linear fractional transformations that map the positive real line to itself.^[25] It is generated by the two parabolic elements T = \begin{pmatrix} 1 & 1 \\ 0 & 1 \end{pmatrix} and U = \begin{pmatrix} 1 & 0 \\ 1 & 1 \end{pmatrix}, where T corresponds to the translation z \mapsto z + 1 and U to the transformation z \mapsto \frac{z}{z + 1}.^[24] The monoid structure arises from matrix multiplication, and every element can be uniquely expressed as a product of these generators, corresponding to words in a two-letter alphabet without relations beyond associativity.^[24] This makes it a free monoid on two generators, isomorphic to the monoid of finite binary strings under concatenation.^[24] It is also isomorphic to the positive braid monoid on three strands.^[26] As a subsemigroup of the modular group \mathrm{PSL}(2, \mathbb{Z}), the dyadic monoid is dense in \mathrm{PSL}(2, \mathbb{R})^+, the connected component of the identity in the group of orientation-preserving Möbius transformations of the upper half-plane.^[25] This density property facilitates its use in the dynamics of the modular group, where products of generators approximate arbitrary elements in the continuous group. Elements of the monoid correspond precisely to reduced words in T and U, avoiding cancellations that occur in the full group presentation.^[24] In applications to dynamical systems, the dyadic monoid plays a role in analyzing horocycle flows on the modular surface \mathbb{H}/\mathrm{PSL}(2, \mathbb{Z}), where its elements provide discrete approximations to continuous horocycle orbits. Similarly, in geodesic approximations, words in the generators yield rational points that densely fill hyperbolic geodesics, aiding the study of continued fraction expansions and Farey sequences.^[27]

Applications

Transformations of the Torus

The modular group \mathrm{PSL}(2, \mathbb{Z}), the projective special linear group over the integers, acts on the space of complex tori through transformations of their underlying period lattices. A complex torus is given by T = \mathbb{C}/\Lambda, where \Lambda is a rank-2 lattice in \mathbb{C}, typically normalized as \Lambda_\tau = \mathbb{Z} \oplus \mathbb{Z}\tau for \tau in the upper half-plane \mathbb{H} = \{\tau \in \mathbb{C} : \operatorname{Im}(\tau) > 0\}. The action of \gamma = \begin{pmatrix} a & b \\ c & d \end{pmatrix} \in \mathrm{SL}(2, \mathbb{Z}) (noting that \mathrm{PSL}(2, \mathbb{Z}) = \mathrm{SL}(2, \mathbb{Z})/\{\pm I\} acts faithfully) sends the basis (1, \tau) to (a\tau + b, c\tau + d), yielding an equivalent lattice \Lambda_{\gamma \tau} up to homothety in \mathbb{C}^*, since tori are unchanged by scaling of the lattice.^[13]^[28] This induces an isomorphism of tori T_\tau \cong T_{\gamma \tau}, preserving the complex structure and making the action a key tool for classifying tori up to biholomorphic equivalence.^[29] The modular parameter \tau parametrizes these tori, and the \mathrm{PSL}(2, \mathbb{Z})-action identifies \tau with \gamma \tau = \frac{a\tau + b}{c\tau + d} for \gamma \in \mathrm{PSL}(2, \mathbb{Z}), grouping equivalent tori into orbits. A complete set of representatives for these orbits lies in the fundamental domain D, a region in \mathbb{H} as described in the section on hyperbolic tessellations. The j-invariant serves as a complete invariant under this action: j(\tau) = 1728 \frac{g_2(\tau)^3}{g_2(\tau)^3 - 27 g_3(\tau)^2}, where g_2 and g_3 are the Eisenstein invariants derived from the Weierstrass \wp-function on the torus, remains constant on orbits since j(\gamma \tau) = j(\tau).^[13]^[28] Thus, two tori T_{\tau_1} and T_{\tau_2} are isomorphic if and only if j(\tau_1) = j(\tau_2).^[29] Over \mathbb{C}, complex tori correspond to elliptic curves via the uniformization theorem, embedding T = \mathbb{C}/\Lambda into \mathbb{P}^2 as a smooth cubic via the map z \mapsto [\wp(z) : \wp'(z) : 1], where \wp is the Weierstrass function. The \mathrm{PSL}(2, \mathbb{Z})-action induces equivalences among these elliptic curves, with isomorphism classes determined by the j-invariant; distinct j-values yield non-isomorphic curves, while isogeny classes over \mathbb{C} coincide with these isomorphism classes due to the absence of non-trivial endomorphisms beyond the lattice structure in generic cases.^[13]^[29] In fact, every complex torus is modularly equivalent under \mathrm{PSL}(2, \mathbb{Z}) to one with \tau \in D, ensuring a canonical representative for each isomorphism class.^[28] Period lattice transformations under the modular group thus provide a discrete symmetry for tori, generated by the translations \tau \mapsto \tau + 1 and inversions \tau \mapsto -1/\tau, reflecting the group's presentation \langle S, T \mid S^2 = (ST)^3 = 1 \rangle. This action not only classifies tori but also underlies the moduli space \mathrm{PSL}(2, \mathbb{Z}) \backslash \mathbb{H}^*, a compact Riemann surface isomorphic to \mathbb{P}^1(\mathbb{C}) via the j-function.^[13]^[28]

Representations in Braid Groups

The modular group \mathrm{PSL}(2, \mathbb{Z}) admits a natural representation as the quotient B_3 / Z(B_3), where B_3 is the 3-strand braid group and Z(B_3) is its center, an infinite cyclic group generated by the full twist \Delta^2 with \Delta = \sigma_1 \sigma_2 \sigma_1. This isomorphism is realized via a surjective homomorphism \phi: B_3 \to \mathrm{PSL}(2, \mathbb{Z}) with kernel Z(B_3), where the Artin generators \sigma_1, \sigma_2 of B_3 satisfy the defining relation \sigma_1 \sigma_2 \sigma_1 = \sigma_2 \sigma_1 \sigma_2. The inverse isomorphism sends the standard generators S = \begin{pmatrix} 0 & -1 \\ 1 & 0 \end{pmatrix} and T = \begin{pmatrix} 1 & 1 \\ 0 & 1 \end{pmatrix} of \mathrm{PSL}(2, \mathbb{Z}) to the classes [\sigma_1 \sigma_2 \sigma_1^{-1} \sigma_2^{-1}] and [\sigma_1^2] in B_3 / Z(B_3), respectively.^[30] The presentation relation (ST)^3 = 1 in \mathrm{[PSL](/page/PSL)}(2, \mathbb{Z}) lifts to the identity \Delta^2 = 1 in the quotient, corresponding directly to the full twist generator of the center Z(B_3).^[31] The pure braid subgroup P_3 \trianglelefteq B_3, which is the kernel of the projection B_3 \to S_3 and generated by elements like A_{12} = \sigma_1^2, A_{13} = \sigma_1 \sigma_2^2 \sigma_1^{-1}, A_{23} = \sigma_2^2, is normalized by the action of B_3, inducing a corresponding action of \mathrm{[PSL](/page/PSL)}(2, \mathbb{Z}) on P_3 via the isomorphism. Connections to the Burau representation arise in low dimensions, particularly for n=3, where the 2-dimensional unreduced Burau representation \beta_t: B_3 \to \mathrm{GL}(2, \mathbb{Z}[t, t^{-1}]) is given explicitly by \beta_t(\sigma_1) = \begin{pmatrix} 1-t & 1 \\ 0 & 1 \end{pmatrix} and \beta_t(\sigma_2) = \begin{pmatrix} t & 0 \\ -1 & 1 \end{pmatrix}, factoring through the quotient to yield a representation of \mathrm{PSL}(2, \mathbb{Z}). The reduced Burau representation for B_3 is faithful, providing a faithful linear representation of the modular group over the Laurent polynomial ring, which captures essential algebraic structure in low-dimensional cases. These representations find applications in knot theory, where closures of braids in B_3 produce knots and links classified up to isotopy by Alexander's theorem, and the induced modular invariants from the quotient map distinguish such objects via representation-theoretic data. For instance, the Alexander polynomial, derived from the Burau representation at t=-1, serves as a modular invariant under the action of \mathrm{[PSL](/page/PSL)}(2, \mathbb{Z}) on the relevant homology groups of knot complements.

Modern Connections

The modular group SL(2, ℤ) acts as the principal symmetry group for classical modular forms, which are holomorphic functions on the upper half-plane transforming under the group's Möbius action as f(\gamma \tau) = (c\tau + d)^k f(\tau) for \gamma = \begin{pmatrix} a & b \\ c & d \end{pmatrix} \in \mathrm{SL}(2, \mathbb{Z}) and weight k \geq 0.^[32] Eisenstein series E_k(\tau), defined for even k \geq 4 via sums over the lattice, are non-cusp modular forms invariant under this full group, generating the Eisenstein subspace of the space of modular forms M_k(\mathrm{SL}(2, \mathbb{Z})).^[32] Cusp forms, which vanish at the cusps of the fundamental domain, form the orthogonal complement, and every modular form for SL(2, ℤ) decomposes uniquely as a linear combination of an Eisenstein series and a cusp form.^[33] The modularity theorem, originally conjectured as the Taniyama-Shimura-Weil conjecture, asserts that every elliptic curve E over \mathbb{Q} is modular, corresponding to a weight-2 newform f (a normalized Hecke eigenform cusp form) on the congruence subgroup \Gamma_0(N) of SL(2, ℤ), where N is the conductor of E, such that the L-function of the curve equals that of the form: L(E, s) = L(f, s).^[34] This equivalence links the arithmetic of elliptic curves, via their Hasse-Weil L-series, to the analytic properties of modular forms in the broader framework of automorphic forms associated to SL(2, ℤ) and its subgroups.^[34] The theorem, proven in stages culminating in full generality by Breuil-Conrad-Diamond-Taylor, underpins connections between elliptic curves and L-functions, with profound implications for the Langlands program.^[34] In physics, monstrous moonshine connects the modular group SL(2, ℤ) to the Monster sporadic simple group M through the vertex operator algebra (VOA) known as the moonshine module V^\natural, whose automorphism group is M and whose graded traces yield the McKay-Thompson series.^[35] These series are hauptmoduln for genus-zero subgroups of SL(2, ℤ), interpolating the J-invariant—a weight-zero modular function for the full group—and dimensions of irreducible representations of M, as observed in the decomposition J(\tau) - 744 = q^{-1} + 196884q + \cdots where 196884 matches the smallest nontrivial M-representation dimension.^[35] The VOA structure, constructed by Frenkel-Lepowsky-Meurman and proven unique up to isomorphism by Dong-Li-Mason, resolves the Conway-Norton conjectures via Borcherds' no-ghost theorem and the Monster Lie algebra.^[35] SL(2, ℤ) also manifests in string theory as the exact S-duality group of type IIB superstring theory, acting on the axion-dilaton scalar \tau = C_0 + i e^{-\phi} (where C_0 is the RR 0-form axion and \phi the dilaton) via fractional linear transformations, interchanging weak and strong coupling and mapping D-branes to F-strings and NS5-branes in multiplets.^[36] This non-perturbative symmetry enhances the classical SL(2, \mathbb{R}) invariance of type IIB supergravity, with Killing spinors transforming as doublets under the duality group to preserve supersymmetry in solutions like the NS5-brane.^[36] In representation theory, SL(2, ℤ) possesses no nontrivial finite-dimensional unitary irreducible representations over \mathbb{C} except those factoring through finite quotients SL(2, \mathbb{Z}/n\mathbb{Z}) for n \geq 1, known as congruence representations.^[37] All such congruence representations are symmetrizable, admitting a basis where the images of the generators S = \begin{pmatrix} 0 & -1 \\ 1 & 0 \end{pmatrix} and T = \begin{pmatrix} 1 & 1 \\ 0 & 1 \end{pmatrix} satisfy \rho(S)^t = \rho(S) and \rho(T) diagonal, facilitating computation of character tables for the quotients via Weil representations and induction from cyclic subgroups.^[37] These characters classify irreducibles, with dimensions bounded by the order of the quotient, and play a role in reconstructing modular tensor categories from the representations.^[37] Recent applications in quantum computing leverage quotients of the modular group in constructing modular fusion categories for topological quantum computation, where SL(2, ℤ)-representations via quantum groups U_q(\mathfrak{sl}_2) at roots of unity enable universal braiding of anyons for fault-tolerant gates.^[38] Post-2020 developments show that finite-dimensional congruence representations of SL(2, ℤ) can be realized as modular representations associated to these categories, allowing explicit construction of 2-semiregular unitary modular tensor categories from levels corresponding to n-quotients, with applications to classifying anyon models for quantum error correction.^[39] Such structures support transversal implementations of quantum gates through braid group actions informed by SL(2, ℤ) symmetries in low-genus surfaces.^[39]

History

Early Developments

The foundations of the modular group trace back to Carl Friedrich Gauss's seminal work on binary quadratic forms. In his 1801 treatise Disquisitiones Arithmeticae, Gauss defined the equivalence of positive definite binary quadratic forms under the action of integer matrices with determinant 1, which corresponds to the transformations now identified as elements of the modular group SL(2, ℤ). This framework allowed him to classify forms up to equivalence and compute class numbers for quadratic fields, laying the groundwork for later group-theoretic interpretations.^[40] Building on Gauss's ideas, Hermann Minkowski advanced the study in the early 1900s through his development of reduction theory within the geometry of numbers. In works such as Geometrie der Zahlen (1910), Minkowski applied lattice theory to quadratic forms, establishing bounds on the number of reduced forms in each equivalence class under the modular group action, thereby providing effective methods to compute class numbers of quadratic fields. His approach emphasized the finite number of reduced representatives, highlighting the modular group's role in bounding the size of ideal classes.^[41] In the 1880s, Henri Poincaré significantly expanded the conceptual scope by introducing automorphic functions and Fuchsian groups. Through papers published in Acta Mathematica (1882–1883), Poincaré defined Fuchsian groups as discrete subgroups of PSL(2, ℝ) acting on the hyperbolic plane, with the modular group serving as the prototypical example due to its action on the upper half-plane. His analysis of fundamental domains and uniformization via these groups connected the modular group to complex function theory, influencing subsequent geometric studies.^[42] The explicit term "modular group" (Modulargruppe) emerged around 1900, attributed to Heinrich Weber and Robert Fricke in their collaborative treatments of elliptic modular functions. In Fricke and Klein's Theorie der elliptischen Modulfunktionen (1890–1892), the group is presented in matrix form as the transformations preserving the j-invariant, formalizing its role in moduli problems. Concurrently, Weber's Lehrbuch der Algebra (1895–1897) discussed the group abstractly in the context of algebraic analysis. Early matrix-based formulations of the modular group also appeared in the late 19th century through the works of Charles Hermite and Camille Jordan. Hermite, in his 1858 memoir on elliptic functions, described the group's generators as linear fractional transformations equivalent to SL(2, ℤ) matrices. Jordan, in Traité des substitutions (1870), analyzed the group as a discrete subgroup of GL(2, ℝ), emphasizing its finite presentation and permutation representations. These contributions bridged number theory with emerging group theory.

Key Advancements

In the late 19th and early 20th centuries, Robert Fricke and Felix Klein made foundational contributions to the structure of the modular group by classifying its subgroups of genus zero and determining fundamental domains for their action on the hyperbolic plane. Their work in Lectures on the Theory of Elliptic Modular Functions established that these subgroups correspond to specific congruence levels, providing a geometric framework for understanding the modular group's tessellations and the associated modular functions. This classification enabled the identification of fundamental regions, such as the standard fundamental domain for PSL(2,ℤ), bounded by the lines Re(z) = ±1/2 and |z| = 1 in the upper half-plane.^[43] In 1967, Morris Newman classified the normal subgroups of finite index in the modular group PSL(2,ℤ), showing that they are precisely the congruence subgroups arising from principal congruence kernels. This result underscored the group's structural simplicity relative to more general Fuchsian groups and highlighted the centrality of congruence subgroups in the normal structure. In the 1920s, Kurt Reidemeister's Schreier method provided tools to describe generators for such subgroups, confirming they are free products of cyclic groups.^[44] In the mid-20th century, A. O. L. Atkin and James Lehner developed the theory of Hecke operators for congruence subgroups like Γ₀(m) of the modular group. Their 1970 paper showed that these operators act diagonally on newforms for Γ₀(m), decomposing spaces of modular forms and revealing eigenvalues for cusp forms. This advancement clarified the arithmetic structure of modular forms for congruence levels, influencing the study of their properties.^[45] Atle Selberg introduced the trace formula in 1956, applying it to the modular group to relate the eigenvalues of the hyperbolic Laplacian on the quotient ℍ/PSL(2,ℤ) to the lengths of closed geodesics. The formula states that for a test function h with Fourier transform ĥ, ∑j ĥ(λ_j) + (1/4π) ∫{-∞}^∞ ĥ(r² - 1/4) (r coth(πr) - 1/r) dr = ∑_{γ} l(γ) h(l(γ)/2) / 2 cosh(l(γ)/2), where λ_j = 1/4 + r_j² are the eigenvalues and the sum over γ is weighted by primitive closed geodesics of length l(γ). This provided the first analytic tool to estimate spectral gaps and multiplicities, proving, for example, that there are infinitely many eigenvalues below 1/4 and bounding the first eigenvalue at approximately 0.004.^[46] During the 1960s and 1970s, Goro Shimura and Robert Langlands developed the theory of automorphic representations, framing modular forms as components of representations of GL(2,ℚ)\GL(2,𝔸). Shimura's 1967 lectures on automorphic functions established the arithmetic properties of these representations, showing that classical modular forms of weight k correspond to holomorphic sections of vector bundles over the modular curve with automorphic factors j(γ,z) = cz + d. Langlands, in collaboration with Hervé Jacquet in 1970, constructed the full space of automorphic forms on GL(2) over the adeles, proving their decomposition into irreducible cuspidal representations and associating L-functions that generalize Dirichlet series for the modular group. This framework unified Hecke theory with representation theory, enabling the study of functoriality and reciprocity for the modular group.^[47]^[48] In the late 20th century, the existence of non-congruence finite index subgroups of the modular group was established, resolving the congruence subgroup problem negatively for PSL(2,ℤ). Works by Irving Reiner (1960) and others demonstrated such subgroups, expanding the study beyond arithmetic levels. Additionally, connections to monstrous moonshine in the 1970s-1980s, via Conway and Norton, linked the modular group to representations of the monster group.^[49]

References

[1]
[PDF] The Modular Group and its Actions
Dec 28, 2013 · 1 Introduction. The projective group of integral two by two matrices of determinant one is called the modular group1 and denoted PSL2(Z). This ...
[2]
[PDF] Introductory Lectures on SL(2,Z) and modular forms.
(1.1) We begin with a definition. The modular group is the subgroup SL(2, Z) of the matrix group SL(2, R) consisting of matrices with integer entries and.
[3]
[PDF] The modular group and some relatives
Most authors consider that, instead of SL2(Z) , the group PSL2(Z) = SL2(Z)/{±I}, where. “P” stands for “projective”, deserves the name of modular group because ...
[4]
https://arxiv.org/pdf/1910.03050.pdf
[5]
Modular Group Gamma -- from Wolfram MathWorld
The group Gamma of all Möbius transformations of the form where a, b, c, and d are integers with ad-bc=1. The group can be represented by the 2×2 matrix.
[6]
[PDF] Algebraic topology - Recitation 7 - Leor Neuhauser
... PSL(2, Z), we want to prove. 4. Page 5. that this map is an isomorphism, implying PSL(2, Z) ≃ Z/2Z ∗ Z/3Z. Elements of the free product w ∈ ⟨α⟩∗⟨β⟩ are ...
[7]
Are the Γ(N) the only normal congruence subgroups of SL2(Z)?
Mar 31, 2011 · So if it is normal its image in SL2(Z)/Γ(N) is a normal subgroup of SL2(Z/NZ). That group is almost simple but not quite. A proper normal ...How does this geometric description of the structure of PSL(2, Z ...The free group $F_2$ has index 12 in SL(2,$\mathbb{Z}$)More results from mathoverflow.net
[8]
Are there noncongruence subgroups (of finite index) of the modular ...
Jan 8, 2013 · For your first question: "Most" 2-generated subgroups of the modular group will have infinite covolume, so they are not congruence subgroups.Concrete examples of noncongruence, arithmetic subgroups of SL(2 ...Why are modular forms (usually) defined only for congruence ...More results from mathoverflow.net
[9]
[PDF] Institute for Mathematical Sciences National University of Singapore ...
. The modular group PSL(2,Z) is the quotient of SL(2,Z) by its center {±I} ... Show that B3 modulo its center is isomorphic with PSL(2,Z). Definition 6 ...
[10]
Representations of the braid group B3 and of SL(2,Z) - MSP
It is known that PSL(2,Z) is isomorphic to the free product Z2 ∗ Z3 of a cyclic group of order 2 with a cyclic group of order 3, and that the iso- morphism can ...
[11]
[PDF] Congruence subgroups of braid groups
The group Z is also the kernel of the symplectic representation ρ : B3 → SL(2,Z) (recalling that Sp(2,Z) = SL(2,Z)). Let. M = ρ(σ1). No element of the coset ...<|control11|><|separator|>
[12]
[PDF] Duke's Theorem and Continued Fractions - arXiv
Feb 20, 2008 · The quotient under this group action SL2(Z)\H has is the fundamental domain represented by the intersection of four sets. {Im(z) > 0}, {|z| < 1} ...
[13]
Hyperbolic fixed points of SL(2,Z) - MathOverflow
Aug 16, 2023 · So every quadratic irrational (in R or in C, depending of what you are transforming) is a fixed point, and this is a complete description. Share.Missing: classification | Show results with:classification
[14]
[PDF] arXiv:2003.12354v4 [math.NT] 4 Mar 2024
Mar 4, 2024 · The Eisenstein series are classified into three cases according to types of γ ∈ Γ, that is, parabolic, elliptic, and hyperbolic Eisenstein.
[15]
[PDF] Basics of binary quadratic forms and Gauss composition
Jun 23, 2014 · Y. ) with M ∈ SL(2,Z). This yields an equivalence relation and splits the binary quadratic forms into equivalence classes. Write ax. 2.
[16]
[PDF] 4.1 Reduction theory - Kimball Martin
We say two forms ax2+bxy+cy2 and Ax2+Bxy+Cy2 are properly equivalent if they satisfy Equation (4.1) for some τ ∈ SL2(Z). In this case we will write ax2 + bxy + ...
[17]
[PDF] Modular Functions and Modular Forms
Theorem 1.27 (Riemann-Hurwitz Formula). Let f : Y → X be a holomorphic ... The Modular Equation for Γ0(N). For any congruence subgroup Γ of Γ(1), the ...
[18]
[PDF] Explicit formulae for linear characters of $\Gamma_0(N)$ - arXiv
Let G be a finite-index subgroup of SL2(Z). Then µ(G) = [PSL2(Z): G]·µ(PSL2(Z)). The expression [PSL2(Z): G] means the index of G in PSL2(Z). For Γ0(N) we ...
[19]
[PDF] Modular Forms and Elliptic Curves
Apr 16, 2021 · For N ∈ N the number ε∞(Γ0(N)) of cusps is. ∑ d|N φ(gcd(d,N/d)) with the sum extending over all positive divisors d of N, and φ denoting the ...
[20]
[PDF] Genus of modular curves
In this essay we will compute the genus for the curves X0(N) for N = 2, 3, 5, 7 and 13, i.e. all N such that N − 1|24. Recall that. Γ0(N) = a b. c d. : c ≡ 0 ...
[21]
[PDF] sl2(z) - keith conrad
Proof of Theorem 1.1. Let G = hS, Ti be the subgroup of SL2(Z) generated by S and T. We will give two proofs that G = SL2(Z), one algebraic and the other ...
[22]
[PDF] (Elliptic Modular Curves) JS Milne
This is an introduction to the arithmetic theory of modular functions and modular forms, with a greater emphasis on the geometry than most accounts. BibTeX ...Missing: rho | Show results with:rho
[23]
[PDF] Notes on the Laplacian and its Eigenfunctions on Bolyai ...
May 28, 2025 · Hyperbolic distance: The geodesic distance between upper-half-plane points z1 = x1 + iy1 and z2 = x2 + iy2 is d(z1,z2) = cosh−1 1 + (x1 ...
[24]
[PDF] Introductory Lectures on SL(2,Z) and modular forms.
In this way, we encounter the classic prototype for a discrete group action, as first considered by Klein and by Poincaré, the modular group Γ(1) ∼= PSL(2, Z).
[25]
[PDF] Fundamental domains for SL 2(Z) and Γ 1. H as homogeneous ...
Oct 21, 2013 · We will see that the simplest quotient of the upper half-plane, SL2(Z)\H, is topologically a sphere with a point missing. However, in the SL2(R)- ...
[26]
[PDF] tessellations of hyperbolic space notes for an ... - Durham University
Fundamental domain for the group Γ. Note that this defines a triangle in H2 with all angles being zero, and with all vertices at the boundary. (“Ideal triangle" ...
[27]
[PDF] Geometric invariants for real quadratic fields - UCLA Mathematics
The Gauss-Bonnet theorem [2,. Thm 10.4.3] gives. (2.8). 1. 2π area(FΓ) = 2(g ... Suppose now that Γ = PSL(2,Z) is the usual modular group. As is well-known ...
[28]
Über die Bestimmung Dirichletscher Reihen durch ihre ... - EUDML
Hecke, E.. "Über die Bestimmung Dirichletscher Reihen durch ihre Funktionalgleichung." Mathematische Annalen 112 (1936): 664-699.
[29]
A fundamental region for Hecke's modular group - ScienceDirect.com
Hecke proved analytically that when λ ≥ 2 or when λ = 2 cos ( π q ) , q ∈ Z, q ≥ 3, then B(λ) = {τ: Im τ > 0, | Re τ| < λ 2 , |τ| > 1} is a fundamental region ...
[30]
[PDF] Matrix Decomposition Problem is Complete for the Average Case ∗
Positive matrices. Call an arbitrary matrix (i.e. an element of. SL2(Z)) positive if it has no negative entries. Pos- itive matrices form a monoid PM = SL2(N).
[31]
[PDF] on two-generator subgroups in sl2(z), sl2(q), and sl2(r)
(b) Let k ∈ Z, k ≥ 2. If M is a matrix from SL2(Z) with nonnegative entries and no elementary operation reduces ∑i,j mij, then either M is the identity matrix ...
[32]
[PDF] arXiv:0810.1265v2 [math.DS] 8 Oct 2008
Oct 8, 2008 · The dyadic monoid is a free monoid generated by two elements g and r, with with r2 = e, and no constraints on g. Alternately, the dyadic monoid ...
[33]
[PDF] THE MINKOWSKI QUESTION MARK AND THE MODULAR GROUP ...
When also passing through the continued fractions, this gives the Minkowski Question Mark function. All of these inter-relationships are presented in this text.
[34]
[PDF] Elliptic Modular Forms and Their Applications
These notes give a brief introduction to a number of topics in the classical theory of modular forms. Some of theses topics are (planned) to be treated.Missing: rho | Show results with:rho<|control11|><|separator|>
[35]
[PDF] 18.783 Elliptic Curves Lecture Note 18 - DSpace@MIT
Apr 18, 2013 · The actions of γ and. −γ are identical, so taking the quotient by PSL2(Z) = SL2(Z)/{±1} yields the same result, but for the sake of clarity we ...
[36]
https://arxiv.org/pdf/hep-th/9912159
[37]
[PDF] The braid group B3 in the framework of continued fractions - arXiv
Aug 5, 2020 · the word problem in the braid group B3. Theorem 1.1. The map ρ : B3 ... extension of the modular group PSL2 (Z) (Proposition 2.5) to define an.
[38]
[PDF] 25 Modular forms and L-series - MIT Mathematics
May 12, 2015 · Theorem 25.31 (Modularity Theorem, formerly the Shimura-Taniyama-Weil8 conjecture). Every elliptic curve E/Q is modular. 25.9 BSD and the parity ...<|separator|>
[39]
[PDF] Modular Forms with Only Nonnegative Coefficients - arXiv
Jul 25, 2025 · Every modular form for SL2(Z) is a linear combination of an Eisenstein series and a cusp form, and every cusp form has both positive and ...<|separator|>
[40]
[PDF] THE SHIMURA-TANIYAMA CONJECTURE - Also re-
The curve E is said to be modular if there exists a cusp form f of weight 2 on Γ0(N) for some N such that L(E,s) = L(f,s). The Shimura-Taniyama conjecture ...
[41]
[PDF] Monstrous Moonshine: The first twenty-five years - arXiv
Apr 14, 2004 · Today we say that there is a vertex operator algebra, called the Moonshine module V ♮, which interpolates between the left and right sides of ( ...
[42]
None
### Summary of the Role of SL(2,Z) in S-duality of Type IIB String Theory
[43]
[PDF] arXiv:2203.15701v1 [math.QA] 29 Mar 2022
Mar 29, 2022 · The linear representations of SL2(Z) arising from modular tensor categories are symmetric and have congruence kernel. Conversely, one may also ...
[44]
[PDF] SL(2,Z) Representations and 2-Semiregular Modular Categories
Mar 4, 2023 · This (real) matrix has non-negative integer entries, so by the Frobenius–Perron theorem [13, Thm. 3.2.1], it has a real, non-negative eigen-.
[45]
[PDF] On the realization of a class of SL(2,Z)-representations - arXiv
Jun 23, 2024 · Recall that the modular group SL(2, Z) is generated by s = 0. 1. −1 0 ! and t = 1 1. 0 1 ! with relations s4 = 1 and (st)3 = s2. The modular ...
[46]
[PDF] The correspondence between binary quadratic forms and quadratic ...
Since the modular group SL2(Z) is a subgroup of GL2(Z), the action above restricts to an action of SL2(Z) on binary quadratic forms, which also yields to an ...
[47]
GAUSS' CLASS NUMBER PROBLEM FOR IMAGINARY ...
A form is called reduced if its associated complex number w lies in the fundamental domain for the modular group SL(2, Z), i.e.,. weSL(2,Z)\$. It is then ...
[48]
[PDF] Introductory Lectures on SL(2,Z) and modular forms.
A real Möbius map is called elliptic if it has one fixed point inside U. It is parabolic if it has one boundary fixed point, hyperbolic if it fixes two boundary ...<|control11|><|separator|>
[49]
https://www.ams.org/journals/bull/1970-76-05/S0273-0979-1970-02788-1/
[50]
[PDF] Normal subgroups of the modular group
In this note we summarize the results of some work on the normal subgroups of the classical modular group r , which is a continuation of the work begun in ...
[51]
Hecke Operators on ... (m). - EUDML
LEHNER, J., and ATKIN, AOL. "Hecke Operators on ... (m).." Mathematische Annalen 185 (1970): 134-160. <http://eudml.org/doc/161948>.Missing: James 1930s
[52]
Automorphic Functions and Number Theory - SpringerLink
Free delivery 14-day returnsBook Title: Automorphic Functions and Number Theory · Authors: Goro Shimura · Series Title: Lecture Notes in Mathematics · Publisher: Springer Berlin, Heidelberg.Missing: 1960s | Show results with:1960s
[53]
[PDF] Automorphic forms on GL(2) Hervé Jacquet and Robert P. Langlands
Page 1. Automorphic forms on GL(2). Hervé Jacquet and Robert P. Langlands. Page 2. Appeared as Lecture Notes in Mathematics 114, Springer-Verlag 1970. Page 3 ...