Introduction#

Fusion categories, a captivating field at the crossroads of mathematics and physics, emerged in the late 1980s from the fertile ground of two-dimensional conformal field theory. Physicists, in their quest to understand the fundamental symmetries of these theories, encountered algebraic structures that went beyond the familiar framework of group representations. It was the seminal work of Gregory Moore and Nathan Seiberg that laid the mathematical foundations for these structures, which they called "fusion rule algebras."

These algebras captured the essential properties of how elementary particles, or more precisely, their corresponding quantum states, "fuse" together. This fusion process is described by a tensor product, a generalization of the familiar tensor product of vector spaces. Just as in group representation theory, where irreducible representations form the building blocks, fusion categories have simple objects that are the fundamental, indecomposable constituents. A key feature that sets fusion categories apart is rigidity, which, in essence, means that for every object, there exists a "dual" object, reminiscent of the dual of a vector space.

The development of fusion categories was also deeply intertwined with the theory of quantum groups, which are non-commutative generalizations of Lie algebras, and with Vaughan Jones's work on subfactor theory in von Neumann algebras. This confluence of ideas revealed that fusion categories provide a unifying language for describing the representation theory of quantum groups and the structure of certain subfactors.

In the early 1990s, the connection to topological quantum field theory (TQFT) became apparent. It was realized that fusion categories provide the algebraic data needed to construct three-dimensional TQFTs, which can be used to compute topological invariants of knots and 3-manifolds. This discovery solidified the importance of fusion categories as a bridge between algebra and topology.

Today, the study of fusion categories is a vibrant and active area of research. Mathematicians and physicists continue to explore their rich structure, with applications ranging from the classification of exotic phases of matter in condensed matter physics to the development of new quantum computing architectures. The ongoing quest to classify fusion categories of a given rank—the number of their simple objects—is a challenging problem that continues to drive innovation in the field.

Definition#

• $k$-linear means that for any two objects $X$ and $Y$ in $\mathcal{C}$, the set of morphisms $\text{Hom}_{\mathcal{C}}(X, Y)$ is a $k$-vector space, and a composition of morphisms is $k$-bilinear.

• Semisimple implies that every object in $\mathcal{C}$ can be expressed as a finite direct sum of simple objects.

• Rigid means that every object $X$ has a left dual ${ }^{\ast}X$ and a right dual $X^{\ast}$. along with the evaluation and coevaluation maps satisfying certain axioms (the snake or zig-zag identities).

• Monoidal refers to the existence of a tensor product functor $\otimes: \mathcal{C} \times \mathcal{C} \to \mathcal{C}$, a unit object $\mathbf{1}$ , and a natural isomorphisms, called associators given by $\alpha_{X, Y, Z}: (X \otimes Y) \otimes Z \to X \otimes (Y \otimes Z)$, and left and right unitors $\lambda_X: \mathbf{1} \otimes X \to X$ and $\rho_X: X \otimes \mathbf{1} \to X$, for all objects $X, Y, Z$ in $\mathcal{C}$, satisfying coherence conditions, called the pentagon and triangle identities.

• The monoidal unit $\mathbf{1}$ is a simple object, meaning $\text{Hom}_{\mathcal{C}}(\mathbf{1}, \mathbf{1}) \cong E$, where $E$ is the ground field. If $k$ is algebraically closed, it is typically assumed that $E \cong k$. However, for a general field $k$, $E = \text{End}_{\mathcal{C}}(\mathbf{1})$ can be a finite field extension of $k$. This field $E$ is the true field of scalars for the category.

Fusion Rule Algebras#

Quantum Groups#

Subfactor Theory#