Centralizers of discrete Temperley-Lieb-Jones subfactors
Jones, McGovern
Discrete, unimodular inclusions of factors $(N\subseteq M, E)$ with $N$ of type $\rm{II}_{1}$ have a natural notion of standard invariant, generalizing the finite index case. When the unitary tensor category of $N$-$N$ bimodules generated by $_{N}L^{2}(M, Ï\circ E)_{N}$ is equivalent to the Temperley-Lieb-Jones category $\text{TLJ}(δ)$, the associated discrete standard invariants are classified in terms of fair and balanced $δ$-graphs. Many examples of these subfactors naturally arise in the context of the Guionnet-Jones-Shlyakhtenko (GJS) construction for graphs. In this paper, we compute the discrete standard invariant of the centralizer subfactor $N\subseteq M^Ï$ for the canonical state $Ï=Ï\circ E$, which is again a discrete subfactor of $\text{TLJ}(δ)$-type. We show that the associated fair and balanced $δ$-graph behaves analogously to a universal covering space of the original fair and balanced $δ$-graph. As an application, we obtain an obstruction to the realization of discrete tracial TLJ-type standard invariants by subfactors of a $\rm{II}_{1}$ factor $M$ in terms of the fundamental group of M.
academic
Centralizers of discrete Temperley-Lieb-Jones subfactors
This paper investigates discrete, unimodular factor inclusions (N⊆M,E), where N is a II1 factor. When the unitary tensor category of N-N bimodules generated by NL2(M,τ∘E)N is equivalent to the Temperley-Lieb-Jones category TLJ(δ), the associated discrete standard invariants can be classified by fair-balanced δ-graphs. The authors compute the discrete standard invariants of the centralizer subfactor N⊆Mϕ (where ϕ=τ∘E), proving that the associated fair-balanced δ-graphs behave like universal covering spaces of the original graphs. As an application, an obstruction condition is obtained for the realization of discrete trace TLJ-type standard invariants in II1 factors.
Extension of finite index subfactor theory: Modern finite index subfactor theory originates from Vaughan Jones's discovery that the set of possible indices for II1 factor inclusions has both discrete and continuous parts. This led to the development of the concept of standard invariants, which have deep connections with category theory, topology, and quantum physics.
Importance of discrete subfactors: Discrete subfactors are natural generalizations of the finite index case, including inclusions of factors into crossed products by discrete group actions. Such subfactors naturally arise in the Guionnet-Jones-Shlyakhtenko (GJS) graph construction.
Challenges in realization problems: Given an abstract standard invariant and a II1 factor M, does there exist a subfactor realizing this standard invariant? This realization problem has been deeply studied in the finite index case, but little is known in the discrete setting.
Computing centralizers: For TLJ-type discrete subfactors, understanding the structure of their centralizer subfactors is necessary
Covering space analogy: Exploring geometric relationships between fair-balanced δ-graphs and their "trace covers"
Providing realization obstructions: Finding necessary conditions for the realization of discrete standard invariants, particularly obstructions related to fundamental groups
Constructed trace cover algorithm: Provided an explicit algorithm for constructing the trace cover Γtr from any fair-balanced δ-graph Γ
Proved covering space properties: Demonstrated that Γtr behaves like a universal covering space of Γ, where "path homotopy" is controlled by edge weights
Established fundamental group obstruction: Proved that the "fundamental group" of the standard invariant T0(M)⊆F(M), providing obstruction conditions for the realization problem
Application to free group factor problem: Applied results to the famous free group factor problem, giving new equivalent characterizations
Given a TLJ-type discrete subfactor N⊆M whose standard invariant is described by a fair-balanced δ-graph (Γ,w,∗), compute the standard invariant of the centralizer N⊆Mϕ, where ϕ=τ∘E.
Step One: Path Graph Construction
Construct graph Γ′ whose vertices correspond to paths in Γ starting from ∗, with edges e~:p1→p2 corresponding to edges e:t(p1)→t(p2), such that p2=p1∗e.
Step Two: Equivalence Relation
Define an equivalence relation ∼tr on vertices of Γ′: v1∼v2 if and only if the corresponding paths have the same endpoint and weight.
Step Three: Quotient GraphΓtr:=Γ′/∼tr
Vertices are denoted [λ,v], representing equivalence classes of paths with weight λ and endpoint v.
Corollary 1.3: Let M be a II1 factor. If q∈/F(M), then there does not exist a discrete, unimodular subfactor N⊆M such that StdInv(N⊆M)≅A−∞,∞q+q−1.
Corollary 1.4: If there exist 0<t<∞ and a discrete, unimodular subfactor N⊆LFt such that its standard invariant is isomorphic to some A−∞,∞δ (δ>2), then LF2≅LF3.
The paper cites core literature in the field, including:
Jones's pioneering work Jon83
Popa's classification theory Pop95a, Pop95b
GJS construction GJS10, GJS11
Discrete subfactor theory JP19
Tensor category theory DCY15
These citations fully demonstrate the authors' deep understanding of the field's development trajectory and the important position of this work in theoretical advancement.