Representation theory of monoidal categories

Representation theory of monoidal categories
Or: Cell theory for monoidal categories
Daniel Tubbenhauer
Part 1: Reps of monoids; Part 2: Reps of algebras
August 2022
Cell theory for monoidal categories Representation theory of monoidal categories August 2022 1 / 7

Where do we want to go?
Groups Monoids
Fusion mon-
oidal cats
Fiat mon-
oidal cats
Group reps Monoid reps
Fusion reps Fiat reps
generalize
“categorify”
rep theory
I Green, Clifford, Munn, Ponizovskiı̆ ∼1940+
+ + many others
Representation theory of (finite) monoids
I Goal Find some categorical analog

Groups Monoids
Fusion mon-
oidal cats
Fiat mon-
oidal cats
Group reps Monoid reps
Fusion reps Fiat reps
generalize
“categorify”
rep theory
I Green, Clifford, Munn, Ponizovskiı̆ ∼1940+
+ + many others
Representation theory of (finite) monoids
I Goal Find some categorical analog
“Categorify”
!
is motivated by

Stasheff polytopes ⇒ I can ignore associators
I Today Cell theory for monoidal categories
I Instead of Rep(G, K) we study Rep(Rep(G, K))
I Examples we discuss Rep(G, K) and S (V ⊗d
|d ∈ N) (“diagram cats”)

The categories in this talk
Categories are monoidal (strict or nonstrict, I won’t be very careful)
Categories are K-linear over some field K
Categories are additive ⊕
Categories are idempotent complete ⊂
⊕
Hom spaces are finite dimensional dimK < ∞
Categories have finitely many indecomposable objects (up to iso)
Not always, but sometimes categories have dualities ?
(rigid, pivotal etc.)

⊕
Everything has a bicategory version
but I completely ignore that!

⊕
Everything has a bicategory version
but I completely ignore that!
Examples
V ec
V ecG /V ecS for a finite group G/monoid S
Rep(G, C), P roj(G, K) or I nj(G, K) for a finite group G
Rep(G, K) for a finite group G sometimes works (details in a sec)
Rep(S, K) for a finite monoid S sometimes (but rarely) works
Categories S (V ⊗d
|d ∈ N) with ⊗-generator V sometimes work (details later)
Quotients of tilting module categories
Projective functor categories CA
Soergel bimodules S bim for finite Coxeter types

Finitary/fiat monoidal cats
Reps
!
matter
simple ! elements
indecomposable ! compounds
I Let S = Rep(G, K)
I S is monoidal
I S is K-linear
I S is additive
I S is idempotent complete
I S has fin dim hom spaces
I S often has infinitely many indecomposable objects
I S has dualities

Reps
!
matter
simple ! elements
I Let S = Rep(G, K)
I S is monoidal
I S is K-linear
I S is additive
I S has dualities
finitary
fiat

Reps
!
matter
simple ! elements
I Let S = Rep(S, K)
I S is monoidal
I S is K-linear
I S is additive
I S often has infinitely many indecomposable objects (even for K = C)
I S has no dualities in general

Z1 ! Z2 ! Z3 !
Z4 ! Z5 !
I Take G = Z/5Z and K = F5, then K[G] ∼
= K[X]/(X5
)
I Rep(G, K) has one simple object Z1 = 1
I Rep(G, K) has five indecomposable objects ⇒ fiat

Z2l :
Z2l+1 :
I Take G = Z/2Z×Z/2Z and K = F2, then K[G] ∼
= K[X, Y ]/(X2
, Y 2
)
I Rep(G, K) has infinitely many indecomposable objects ⇒ not fiat

Z2l :
Z2l+1 :
= K[X, Y ]/(X2
, Y 2
)
Theorem (Higman ∼1954)
Rep(G, K) is fiat if and only if either
(a) char(K) does not divide |G|
or
(b) char(K) = p divides |G| and the p-Sylow subgroups of G are cyclic

Z2l :
Z2l+1 :
= K[X, Y ]/(X2
, Y 2
)
or
Examples and nonexamples
Rep(S3, F2), Rep(Dodd, F2) are fiat
Rep(S4, F2), Rep(Deven, F2) are not fiat
Blue circle = cyclic subgroups, green = 2-Sylows

Z2l :
Z2l+1 :
= K[X, Y ]/(X2
, Y 2
)
or
Together with P roj(G, K) and I nj(G, K) (these are always fiat )
Higman’s theorem provides many examples of fiat categories
A Higman theorem for monoids is widely open
but one shouldn’t expect it too be very nice, e.g.
Tn has finite representation type over C ⇔ n ≤ 4

Reps
!
matter
simple ! elements
I Let S = S (V ⊗d
|d ∈ N) (+ K-linear + ⊕ + ⊂
⊕ ) for some nice V
I S is monoidal
I S is K-linear
I S is additive
I S has fin dim hom spaces (
I S has dualities ( ) depends but is easy to check

Reps
!
matter
simple ! elements
I Let S = S (V ⊗d
|d ∈ N) (+ K-linear + ⊕ + ⊂
I S is monoidal
I S is K-linear
I S is additive
Almost examples
Temperley–Lieb (TL), Brauer or Deligne categories
and other diagram categories in the same spirit
Catch These usually have infinitely many indecomposable objects
⇒ truncate these appropriately

Reps
!
matter
simple ! elements
I Let S = S (V ⊗d
|d ∈ N) (+ K-linear + ⊕ + ⊂
I S is monoidal
I S is K-linear
I S is additive
Example/Theorem (Alperin, Kovács ∼1979)
“Finite TL”, i.e. V any simple of G = SL2(Fpk ) over characteristic p
S (V ⊗d
|d ∈ N) is fiat , e.g. p = 5, K = F5, k = 2, V = (F25)2
:
simples in Rep(G, K):
indecomposables in Rep(G, K):
indecomposables in S (V ⊗d
|d ∈ N):

Reps
!
matter
simple ! elements
I Let S = S (V ⊗d
|d ∈ N) (+ K-linear + ⊕ + ⊂
I S is monoidal
I S is K-linear
I S is additive
Example/Theorem (folklore)
V any 2d simple of a finite group G
S (V ⊗d
|d ∈ N) is finitary ,
e.g. K = F2, V the two dim simple of G = D6:
|d ∈ N):

Reps
!
matter
simple ! elements
I Let S = S (V ⊗d
|d ∈ N) (+ K-linear + ⊕ + ⊂
I S is monoidal
I S is K-linear
I S is additive
Algebraic modules à la Alperin
provide many examples of finitary/fiat “diagram lookalike cats”
The state of the arts for algebraic modules is roughly the same as for algebraic numbers:
there are some results, but not so many
In the monoid case next to nothing is known

Reps
!
matter
simple ! elements
I Let S = S (V ⊗d
|d ∈ N) (+ K-linear + ⊕ + ⊂
I S is monoidal
I S is K-linear
I S is additive
Example/Theorem (Craven ∼2013)
V any simple of M11 in characteristic 2
S (V ⊗d
|d ∈ N) is finitary ,
e.g. V the 10 dim simple of G = M11:
|d ∈ N):
There are many similar results known, but they all look a bit random, e.g.

Cells in monoidal cats
The categorical cell orders and equivalences for the set of indecomposables B:
Left, right and two-sided cells (a.k.a. L, R and J-cells): equivalence classes
I H-cells = intersections of left and right cells
I Slogan Cells measure information loss

I H-cells = intersections of left and right cells
I Slogan Cells measure information loss
Green cells in categories
B = {X, Y, Z, ...} set of indecomposables of a finitary monoidal category S
⊂
⊕ = is direct summand of

I Get monoidal semicategories SJ , SH by killing higher order terms
I I tell you later which ones are “idempotent”

Example (Rep(S3, C))
Indecomposable objects Z1
∼
= 1 ! , Z2 ! , Z3 !
1 ⊂
⊕ ⊗ ⇒ is in the lowest cell
1 ⊂
Only one cell

∼
= 1 ! , Z2 ! , Z3 !
1 ⊂
1 ⊂
Only one cell
Example (Rep(G, C))
1 ⊂
⊕ Z ⊗ Z∗
⇒ Z is in the lowest cell
Only one cell

∼
= 1 ! , Z2 ! , Z3 !
1 ⊂
1 ⊂
Only one cell
Example (Rep(G, C))
1 ⊂
⊕ Z ⊗ Z∗
Only one cell
Example (semisimple + dual (replaces ( )−1
))
1 ⊂
⊕ Z ⊗ Z∗
Only one cell

Example (S (V ⊗d
|d ∈ N) for the 2d simple S3 rep over F2)
∼
= 1 ! , Z2 ! , Z3 = P(1)
⊗ ∼
= Z3 ⊕ Z3
⊗ Z3
∼
= ⊕ Z3
Z3 ⊗ Z3
∼
= ⊕
Two cells
Z2, Z3
1
Jt
Jb
SH
∼
= ??
SH
∼
= V ec

Example (S (V ⊗d
∼
= 1 ! , Z2 ! , Z3 = P(1)
⊗ ∼
= Z3 ⊕ Z3
⊗ Z3
∼
= ⊕ Z3
Z3 ⊗ Z3
∼
= ⊕
Two cells
Z2, Z3
1
Jt
Jb
SH
∼
= ??
SH
∼
= V ec
In general, for S ⊂ Rep(G, K)
the top J cell is the cell of projectives

Example (S (V ⊗d
∼
= 1 ! , Z2 ! , Z3 = P(1)
⊗ ∼
= Z3 ⊕ Z3
⊗ Z3
∼
= ⊕ Z3
Z3 ⊗ Z3
∼
= ⊕
Two cells
Z2, Z3
1
Jt
Jb
SH
∼
= ??
SH
∼
= V ec
In general, for S ⊂ Rep(G, K)
the top J cell is the cell of projectives
Warning
For S ⊂ Rep(S, K)
the top J cell is usually not the cell of projectives
Dualities are helpful

Example/theorem (folklore)
S (V ⊗d
|d ∈ N) for “finite TL” over Fpk
There are (k + 1) cells
Zpk −1, ..., Z2pk −2
.
.
.
Zp3−1, ..., Zp4−2
Zp2−1, ..., Zp3−2
Zp−1, ..., Zp2−2
Z0 = 1, ..., Zp−2
Jt
J3
J2
J1
Jb
SH
∼
= V erpk
SH
∼
= V erp3
SH
∼
= V erp2
SH
∼
= V erp
SH
∼
= V er
where V er is the semisimplification of SL2(Fp) tilting modules
and the other SH are “higher” Verlinde cats

Example (projective functors)
A some reasonable algebra, 1 = e1 + e2 primitive orthogonal idempotents
CA finitary monoidal category of projective functors + id functor
There are 2 cells

Example (Soergel bimodules)
S bim is fiat monoidal category for finite Coxeter type
Cells = p cells
For type B2 (dihedral group D4) one has e.g.

Reps of monoidal cats
Frobenius: act on linear spaces
Schur: act on projective spaces
Varying the source/target gives slightly different theories
I Start with examples In a sec
I Choose the type of categories you want to represent Finitary/fiat monoidal
I Choose the type of categories you want as a target Finitary
I Build a theory Depends crucially on the setting

Frobenius: act on linear spaces
Schur: act on projective spaces
Varying the source/target gives slightly different theories
I Start with examples In a sec
I Choose the type of categories you want to represent Finitary/fiat monoidal
I Choose the type of categories you want as a target Finitary
I Build a theory Depends crucially on the setting
Some flavors, varying source/target
Categorical reps of groups (subfactors, fusion cats, etc.)
à la Jones, Ocneanu, Popa, others ∼1990
Categorical reps of Lie groups/Lie algebras
à la Chuang–Rouquier, Khovanov–Lauda, others ∼2000
Categorical reps of algebras ( abelian , tensor cats, etc.)
à la Etingof, Nikshych, Ostrik, others ∼2000
Categorical reps of monoids/algebras ( additive , finitary/fiat monoidal cats, etc.)
à la Mazorchuk, Miemietz, others ∼2010

I Let S = Rep(G, K)
I The regular cat module M: S → End(S ):
M //
f

M ⊗
f ⊗

N // N ⊗
I The decategorification is an N -module
Example (G = S3, K = C)
Z1
∼
= 1 ! , Z2 ! , Z3 !
[M(Z1)] !


1 0 0
0 1 0
0 0 1

 , [M(Z2)] !


0 1 0
1 1 1
0 1 0

 , [M(Z3)] !


0 0 1
0 1 0
1 0 0



I Let K ⊂ G be a subgroup
I Rep(K, K) is a cat module of Rep(G, K) via
M(K, 1) = ResG
K ⊗ : Rep(G, K) → End Rep(K, K)

,
M //
f

ResG
K (M) ⊗
ResG
K (f )⊗

N // ResG
K (N) ⊗
I The decategorifications are N -modules
Example (G = S3, K = S2, K = C, M = M(K, 1))
→ , → ⊕ , →
[M(Z1)] !

1 0
0 1

, [M(Z2)] !

1 1
1 1

, [M(Z3)] !

0 1
1 0


I Let ϕ ∈ H2
(K, C∗
), and M(K, ϕ) be the category of projective K-modules
with Schur multiplier ϕ, i.e. a vector spaces V with ρ: K → End(V ) such that
ρ(g)ρ(h) = ϕ(g, h)ρ(gh), for all g, h ∈ K
I Note that M(K, 1) = Rep(K) and
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
I M(K, ϕ) is also a cat module of S :
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
I The decategorifications are N -modules – the same ones from before!

I Let ϕ ∈ H2
(K, C∗
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
M(K, ϕ) are solutions to equations on the Grothendieck level
and
the categorical level

I Let ϕ ∈ H2
(K, C∗
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
M(K, ϕ) are solutions to equations on the Grothendieck level
and
the categorical level
Goal
Find some setting where M(K, ϕ) naturally fit into
(I really like them!)

I Let ϕ ∈ H2
(K, C∗
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
Source/target
I want finitary/fiat categories to act
My target categories are finitary

I Let ϕ ∈ H2
(K, C∗
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
Source/target
Decat
M is called transitive if it is nonzero and is generated by any nonzero X

I Let ϕ ∈ H2
(K, C∗
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
Source/target
Decat
Cat
M is called simple (transitive) if there are no nontrivial S -stable ideals

I Let ϕ ∈ H2
(K, C∗
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
Source/target
Decat
Cat
Example (Rep(S3, C) and M = M(S3, φ))
M is transitive because T = Z1 ⊕ Z2 ⊕ Z3 has a connected action matrix
T !


1 1 1
1 2 1
1 1 1

 !

I Let ϕ ∈ H2
(K, C∗
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
Source/target
Decat
Cat
M is transitive because T = Z1 ⊕ Z2 ⊕ Z3 has a connected action matrix
T !


1 1 1
1 2 1
1 1 1

 !
M is simple because its transitive and hom spaces are boring

I Let ϕ ∈ H2
(K, C∗
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
Theorem (Mazorchuk–Miemietz ∼2014)
In the correct framework
cat reps satisfy a (weak) Jordan–Hölder theorem wrt simple cat reps
(weak = get transitive subquotients and kill ideals)
Goal
For fixed S , find the periodic table of simple cat reps

I Let ϕ ∈ H2
(K, C∗
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
Theorem (Ocneanu ∼1990, folklore)
Completeness
All simples of Rep(G, C) are of the form M(K, ϕ).
Non-redundancy
We have M(K, ϕ) ∼
= M(K0
, ϕ0
) ⇔ the subgroups and cocycles are conjugate

I Let ϕ ∈ H2
(K, C∗
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
Completeness
Non-redundancy
= M(K0
, ϕ0
Example (G = S3 at the top, G = S4 at the bottom)

I Let ϕ ∈ H2
(K, C∗
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
Example/theorem (Etingof, Ostrik ∼2003)
The Hopf algebra T = hg, z|gn
= 1, zn
= 0, gz = ζzgi
for a primitive complex nth root of unity ζ ∈ C
T is the Taft algebra (a well known but nasty example in Hopf algebras)
Rep(T, C) is fiat monoidal with two cells
Rep(T, C) has infinitely many simple reps
but only finitely many Grothendieck classes of simple reps
There are infinity many twists of the actions

Cells and reps of monoidal cats
Clifford, Munn, Ponizovskiı̆ ∼1940+
+ H-reduction
There is a one-to-one correspondence

simples with
apex J (e)

one-to-one
←
−
−
−
−
→

simples of (any)
H(e) ⊂ J (e)

Reps of monoids are controlled by H(e) cells
I We already have cell theory in monoidal cats
I Goal Find an H-reduction in the monoidal setup

+ H-reduction

simples with
apex J (e)

one-to-one
←
−
−
−
−
→

simples of (any)
H(e) ⊂ J (e)

Duflo involution
D = D(L) is Duflo if it satisfies the universal property:
∃ γ : D → 1 such that
Fγ : FD → F right splits (Fγ ◦ s = idF ) for all F ∈ L
“Duflo involution = nonnegative pseudo idempotent”
Having a Duflo involution implies that L has a
nonnegative pseudo idempotent
= coefficients from N wrt the basis of classes of indecomposables

+ H-reduction

simples with
apex J (e)

one-to-one
←
−
−
−
−
→

simples of (any)
H(e) ⊂ J (e)

Duflo involution
Example (Rep(G, C))
The unique Duflo involution is 1

+ H-reduction

simples with
apex J (e)

one-to-one
←
−
−
−
−
→

simples of (any)
H(e) ⊂ J (e)

Duflo involution
Example (Rep(G, C))
The unique Duflo involution is 1
Example (S bim of dihedral type, n odd)
pseudo idempotents (left) and nonnegative pseudo idempotent (right):
(Recall from the exercises that b12 − b1212± was a pseudo idempotent)

+ H-reduction

simples with
apex J (e)

one-to-one
←
−
−
−
−
→

simples of (any)
H(e) ⊂ J (e)

S (V ⊗d
Zpk −1, ..., Z2pk −2
.
.
.
Zp3−1, ..., Zp4−2
Zp2−1, ..., Zp3−2
Zp−1, ..., Zp2−2
Z0 = 1, ..., Zp−2
Jt
J3
J2
J1
Jb
SH
∼
= V erpk
SH
∼
= V erp3
SH
∼
= V erp2
SH
∼
= V erp
SH
∼
= V er
The Steinberg modules Zpj −1 are the Duflo involutions

In spirit of Clifford, Munn, Ponizovskiı̆ ∼1940+
+ H-reduction
There is a one-to-one correspondence (currently only proven in the fiat case)

simples with
apex J

one-to-one
←
−
−
−
−
→

simples of
SH

Reps are controlled by the SH categories
I Each simple has a unique maximal J where having a pseudo idempotent is
replaced by Duflo involutions Apex
I This implies (smod means the category of simples):
S -smodJ ' SH-smod

+ H-reduction

simples with
apex J

one-to-one
←
−
−
−
−
→

simples of
SH

S -smodJ ' SH-smod
Example (Rep(G, C))
H-reduction is not really a reduction and we need Ocneanu’s classification

+ H-reduction

simples with
apex J

one-to-one
←
−
−
−
−
→

simples of
SH

S -smodJ ' SH-smod
Example (Rep(G, C))
H-reduction is not really a reduction and we need Ocneanu’s classification
Example (S bim)
H-reduction reduces the classification problem a lot
but one needs extra work to complete it (the SH are complicated)

Reps
!
matter
simple ! elements
I Let S = Rep(G, K)
I S is monoidal
I S is K-linear
I S is additive
I S has dualities
finitary
fiat
Z1 ! Z2 ! Z3 !
Z4 ! Z5 !
= K[X]/(X5
)
finitary
fiat
Z2l :
Z2l+1 :
= K[X, Y ]/(X2
, Y 2
)
finitary
fiat
Z2l :
Z2l+1 :
= K[X, Y ]/(X2
, Y 2
)
finitary
fiat
or
Reps
!
matter
simple ! elements
I Let S = S (V ⊗d
|d ∈ N) (+ K-linear + ⊕ + ⊂
I S is monoidal
I S is K-linear
I S is additive
finitary
fiat
Almost examples
S (V ⊗d
Zpk −1, ..., Z2pk −2
.
.
.
Zp3−1, ..., Zp4−2
Zp2−1, ..., Zp3−2
Zp−1, ..., Zp2−2
Z0 = 1, ..., Zp−2
Jt
J3
J2
J1
Jb
SH
∼
= V erpk
SH
∼
= V erp3
SH
∼
= V erp2
SH
∼
= V erp
SH
∼
= V er
I Let ϕ ∈ H2
(K, C∗
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
Completeness
Non-redundancy
= M(K0
, ϕ0
I Let ϕ ∈ H2
(K, C∗
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
= 1, zn
= 0, gz = ζzgi
+ H-reduction

simples with
apex J

one-to-one
←
−
−
−
−
→

simples of
SH

S -smodJ ' SH-smod
There is still much to do...

Reps
!
matter
simple ! elements
I Let S = Rep(G, K)
I S is monoidal
I S is K-linear
I S is additive
I S has dualities
finitary
fiat
Z1 ! Z2 ! Z3 !
Z4 ! Z5 !
= K[X]/(X5
)
finitary
fiat
Z2l :
Z2l+1 :
= K[X, Y ]/(X2
, Y 2
)
finitary
fiat
Z2l :
Z2l+1 :
= K[X, Y ]/(X2
, Y 2
)
finitary
fiat
or
Reps
!
matter
simple ! elements
I Let S = S (V ⊗d
|d ∈ N) (+ K-linear + ⊕ + ⊂
I S is monoidal
I S is K-linear
I S is additive
finitary
fiat
Almost examples
S (V ⊗d
Zpk −1, ..., Z2pk −2
.
.
.
Zp3−1, ..., Zp4−2
Zp2−1, ..., Zp3−2
Zp−1, ..., Zp2−2
Z0 = 1, ..., Zp−2
Jt
J3
J2
J1
Jb
SH
∼
= V erpk
SH
∼
= V erp3
SH
∼
= V erp2
SH
∼
= V erp
SH
∼
= V er
I Let ϕ ∈ H2
(K, C∗
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
Completeness
Non-redundancy
= M(K0
, ϕ0
I Let ϕ ∈ H2
(K, C∗
⊗: M(K, ϕ) M(K, ψ) → M(K, ϕψ)
Rep(G, C) M(K, ϕ)
ResG
K Id
−
−
−
−
−
−
→ Rep(K) M(K, ϕ)
⊗
−
→ M(K, ϕ)
= 1, zn
= 0, gz = ζzgi
+ H-reduction

simples with
apex J

one-to-one
←
−
−
−
−
→

simples of
SH

S -smodJ ' SH-smod
Thanks for your attention!

Representation theory of monoidal categories

More Related Content

Similar to Representation theory of monoidal categories (20)

More from Daniel Tubbenhauer (10)

Recently uploaded (20)

Representation theory of monoidal categories