nLab Tambara functor

Contents

Contents

Idea

Tambara functors are algebraic structures similar to Mackey functors, but with multiplicative norm maps as well as additive transfer maps, and a rule governing their interaction. They were introduced by Tambara, as TNR-functors, to encode the relationship between Trace (additive transfer), Norm (multiplicative transfer) and Restriction maps in the representation theory and cohomology theory of finite groups (Tam93).

In the GG-equivariant context for a finite group GG, the role of abelian groups in non-equivariant algebra is played by Mackey functors. The category of Mackey functors is a closed symmetric monoidal category with symmetric monoidal product, the box product. In addition to the expected generalization of commutative rings to simply commutative monoids for the box product, there is a poset of generalizations of the notion of commutative rings to the GG-equivariant context: the incomplete Tambara functors. These interpolate between Green functors, the ordinary commutative monoids for the box product, and Tambara functors. The distinguishing feature for [incomplete] Tambara functors is the presence of certain multiplicative transfer maps, called norm maps. For a Green functor, we have no norm maps; for a Tambara functor, we have norm maps for any pair of subgroups HKH \subset K of GG. (Hill 17)

A Mackey functor is represented by a “GG-equivariant Eilenberg–MacLane spectrum”, … a Tambara functor is represented by a commutative “GG-equivariant Eilenberg-MacLane ring spectrum” (AB15)

Definition

In parts

Given a sequence of maps of finite GG-sets XgYhZX \xrightarrow{g} Y \xrightarrow{h} Z, we define distributor diagram

Δ(g,h)=(XpAqBrZ) \Delta(g,h) = \left( X \xleftarrow{p} A \xrightarrow{q} B \xrightarrow{r} Z \right)

by setting

B{(z,s)zZ,s:h 1(z)X, s.t. gs=id h 1(z)} B \coloneqq \left\{ (z,s) \mid z \in Z, s\colon h^{-1}(z) \rightarrow X, \text{ s.t. } gs = \mathrm{id}_{h^{-1}(z)} \right\}
AY× ZB A \coloneqq Y \times_Z B

and setting p(y,s)=s(y)p(y,s) = s(y), q(y,s)=(h(y),s)q(y,s) = (h(y),s), and r(z,s)=zr(z,s) = z.

Definition

A semi-Tambara functor for the group GG is the data of

  • a GG-coefficient system of sets RR and

  • two GG-semi-Mackey functors (R,T)(R,T) and (R,N)(R,N) with coefficient system RR,

subject to the distributivity condition that, for all distributors Δ(f,g)=(XpAqArZ)\Delta(f,g) = (X \xleftarrow{p} A \xrightarrow{q} A \xrightarrow{r} Z), we have

N gT f=T rN qR p. N_g T_f = T_r N_q R_p.

A semi-Tambara functor is a Tambara functor if (R,T)(R,T) is a Mackey functor.

We often refer to the maps R fR_f as restrictions, T fT_f as transfers, and N fN_f as norms.

Categorically

We define a 2-category to have

  • objects: finite GG sets

  • morphisms: bispan diagrams IpXfYqJI \xleftarrow{p} X \xrightarrow{f} Y \xrightarrow{q} J in finite GG-sets

  • 2-cells: isomorphisms of bispan diagrams

The identity morphisms and 2-cells are as one would expect, and composition is defined by the outer bispan diagram

whose blue inner diagram is defined by the distributor

Δ(g,f)=(X× JYpY× ZBqBrZ). \Delta(g,f) = \left( X \times_J Y \xleftarrow{p} Y \times_Z B \xrightarrow{q} B \xrightarrow{r} Z \right).

Definition

Let 𝒞\mathcal{C} by an ∞-category admitting finite products. Then, a GG-semi-Tambara functor valued in 𝒞\mathcal{C} is a product preserving functor Bispan(𝔽 G)𝒞\mathrm{Bispan}(\mathbb{F}_G) \rightarrow \mathcal{C}. A GG-semi-Tambara functor is a Tambara functor if its underlying additive semi-Mackey functor is a Mackey functor.

In this formalism, given a semi-Tambara functor XX, restriction R K H:X(H)X(K)R_K^H\colon X(H) \rightarrow X(K) is implemented by functoriality under the bispan

G/HG/K=G/K=G/K, G/H \leftarrow G/K = G/K = G/K,

norms by

G/K=G/KG/H=G/H, G/K = G/K \rightarrow G/H = G/H,

and transfers by

G/K=G/K=G/KG/H. G/K = G/K = G/K \rightarrow G/H.

Examples

(under construction…)

Burnside rings, representation rings, zeroth stable homotopy group of a genuine equivariant E E_{\infty}-ring.

the homotopy category of Eilenberg MacLane commutative ring spectra is equivalent to the category of Tambara functors. (Ull13)

Some other examples are related to Witt-Burnside functors, Witt rings in the sense of Dress and Siebeneicher.

Additive completion of semi-Tambara functors

The following is Proposition 13.23 of Strickland 12.

Proposition

The additive completion of semi-Mackey functors underlies a left adjoint sTamb(𝒞)Tamb(𝒞)\mathrm{sTamb}(\mathcal{C}) \rightarrow \mathrm{Tamb}(\mathcal{C}) to the forgetful functor.

The Borel construction

Given SS a semiring with GG-action through homomorphisms, we have a coefficient system S S^{\bullet} whose [G/H][G/H]-value is the invariants S HS^H. We give this the additional structure of a semi-mackey functor by first using the semiring isomorphism

S HHom G(G/H,S) S^H \simeq \mathrm{Hom}^G(G/H,S)

under the pointwise semiring structure, then given a map of transitive GG-sets r:G/KG/Hr\colon G/K \rightarrow G/H, writing

T K H(f)(x)= r(y)=xf(y). T_K^H(f)(x) = \sum_{r(y) = x} f(y).

We then give this the structure of a Tambara functor along q:G/KG/Hq\colon G/K \rightarrow G/H by the formula

N K H(f)(x)= q(y)=xf(y) N_K^H(f)(x) = \prod_{q(y) = x} f(y)

In particular, we may view T K HT_K^H as being defined by a “left Kan extension” formula and N K HN_K^H via “right Kan extension.”

Note that restricting to the value on G/eG/e yields a functor U:sTamb G(𝒞)sRing G(𝒞)U\colon \mathrm{sTamb}_{G}(\mathcal{C}) \rightarrow \mathrm{sRing}_{G}(\mathcal{C}). The following is Example 6.3 of [Strickland 12]

Proposition

The Borel construction is right adjoint to U:sTamb G(𝒞)sRing G(𝒞)U\colon \mathrm{sTamb}_{G}(\mathcal{C}) \rightarrow \mathrm{sRing}_{G}(\mathcal{C}).

Representation Tambara functors

Given HGH \subset G a subgroup, note that (isomorphism classes of) HH-representations correspond with GG-equivariant vector bundles over [G/H][G/H]:

Vect k G([G/H])Rep H. \mathrm{Vect}_k^{G}([G/H]) \simeq \mathrm{Rep}^H.

Restriction gives these the structure of a coefficient system; we may give these a semiring structure under (,)(\oplus,\otimes). Moreover, we lift this to a semi-Tambara structure with transfers given by induction

T K H(V) x r(y)=xV y T_K^H(V)_x \simeq \bigoplus_{r(y) = x} V_y

and norms given by tensor-induction

N K H(V) x q(y)=xV y. N_K^H(V)_x \simeq \bigotimes_{q(y) = x} V_y.

The Representation Tambara functor is the additive completion of this.

Burnside Tambara functors

Let 𝒜(S)\mathcal{A}(S) be the Grothendieck group

𝒜(S)K 0((𝔽 G) /S). \mathcal{A}(S) \coloneqq K_0 \left((\mathbb{F}_G)_{/S} \right).

This becomes a coefficient system under precomposition. Moreover, colimits in 𝔽 G\mathbb{F}_G are universal, so precomposition along a map of finite GG-sets STS \rightarrow T possesses a both a left and right adjoint; the left adjoint to [G/K][G/H][G/K] \rightarrow [G/H] is called induction and the right adjoint is called coinduction. The Burnside Tambara functor has coefficient system 𝒜()\mathcal{A}(-), transfers given by induction, and norms given by coinduction.

An explicit example

Let G=C 2={e,σ}G = C_2 = \{ e, \sigma \}. Then, a C_2-Mackey functor consists of the data

  • an Abelian group XX with C 2C_2-action,
  • an Abelian group YY,
  • a C 2C_2-equivariant restriction homomorphism R:YXR\colon Y \rightarrow X (under trivial action on YY), and
  • a C 2C_2-equivariant transfer homomorphism T:XYT\colon X \rightarrow Y,

subject to the double coset formula

RT(a)=a+σa. RT(a) = a + \sigma a.

Thus, a C 2C_2-Tambara functor consists of * a semiring XX with C 2C_2-action, * a semiring YY, * a C 2C_2-equivariant semiring homomorphism R:YXR\colon Y \rightarrow X (under trivial action on YY), * a C 2C_2-equivariant additive map T:XYT\colon X \rightarrow Y, and * a C 2C_2-equivariant multiplicative map N:XYN\colon X \rightarrow Y,

subject to the double coset formulas

  • RT(a)=a+σaRT(a) = a + \sigma a

  • RN(a)=aσaRN(a) = a \cdot \sigma a

and the distributivity formula

  • T(aR(b))=T(a)b.T(a R(b)) = T(a) b.

We may explicitly define a C 2C_2-Tambara funcgtor by underlying coefficient system semiring X=[α]/α 2X = \mathbb{Z}[\alpha] / \alpha^2, Y=[β,γ]/(β 2,βγ,γ 2,2γ)Y = \mathbb{Z}[\beta, \gamma] / (\beta^2, \beta \gamma, \gamma^2, 2\gamma), under trivial C 2C_2-action and restriction map

R(i+jβ+kγ)=i+2jα. R(i + j\beta + k\gamma) = i + 2j\alpha.

We may give this a Tambara structure by transfer

T(i+jα)=2i+jβ T(i + j\alpha) = 2i + j\beta

and norm

N(i+jα)=i 2+ijβ+j 2γ. N(i + j\alpha) = i^2 + ij\beta + j^2 \gamma.

References

Originally,

  • Daisuke Tambara, On multiplicative transfer, Comm. Algebra 21 (1993), no. 4, 1393–1420 (pdf).

Other references in homotopy theory,

On variations of Tambara functors,

Relationship with Witt vectors,

In algebra,

  • Hiroyuki Nakaoka?, A generalization of The Dress construction for a Tambara functor, and polynomial Tambara functors, (arXiv:1012.1911)

  • Hiroyuki Nakaoka?, Ideals of Tambara functors (arXiv:1101.5982),

  • Hiroyuki Nakaoka?, On the fractions of semi-Mackey and Tambara functors (arXiv:1103.3991),

  • Hiroyuki Nakaoka?, Biset transformations of Tambara functors (arXiv:1105.0714),

  • Hiroyuki Nakaoka?, Spectrum of the Burnside Tambara functor on a cyclic pp-group (arXiv:1301.1453).

  • Maxine Calle, Sam Ginnett, The Spectrum of the Burnside Tambara Functor of a Cyclic Group, (arXiv:2011.04729)

  • Noah Wisdom, A classification of C p nC_{p^n}-Tambara fields, (arXiv:2409.02966)

  • David Chan?, David Mehrle, J.D. Quigley, Ben Spitz?, Danika Van Niel?, On the Tambara Affine Line (arXiv:2410.23052)

In derived algebra,

In higher algebra,

Last revised on December 21, 2024 at 16:00:57. See the history of this page for a list of all contributions to it.