Integral Transforms for Finite Gauge Theory
Jackson Van Dyke
TL;DR
The paper addresses the quantization of $\\pi$-finite spaces in dimension three and analyzes how symmetries act on the resulting TQFTs at two levels: projectively on the full phase space and linearly on a fixed Lagrangian. It introduces a twice-categorified analogue of Blattner-Kostant-Sternberg kernels to describe projective actions of orthogonal symmetries on the Dijkgraaf-Witten-type theory $\\sigma_{BL}^3$, while Aut$(L)$ yields coherent linear actions, linking to metaplectic-type phenomena for finite abelian metric groups. The framework relies on a $3$-functor formalism for spans of $\\pi$-finite groupoids, enabling canonical bimodule-categorical descriptions of these symmetries and their obstructions, including an anomaly captured by a $4$-dimensional bulk theory in the orthogonal case. Connections are drawn to Crane-Yetter, Reshetikhin-Turaev/Turaev-Viro theories, and relative Langlands-type structures, illustrating how gauging symmetries of finite phase-space data produces enriched $3$-manifold invariants with potential physical and mathematical applications. Overall, the work establishes a precise correspondence between higher-categorical quantization of finite spaces and the interplay of linear vs. projective symmetry actions in $3$-dimensional topological gauge theories.
Abstract
This paper shows that quantization of $π$-finite spaces, as a functor out of a higher category of spans, is equivariant in two ways: Symmetries of a given polarization/Lagrangian always induce coherent symmetries of the quantization. On the other hand, symmetries of the entire phase space a priori only induce projective symmetries, with an invertible once-categorified theory, the anomaly theory, encoding the projectivity. We give projective symmetries of three-dimensional finite gauge theories a concrete description via a twice-categorified analogue of Blattner-Kostant-Sternberg kernels and the associated integral transforms, such as the Fourier transform. This establishes an analogy between certain instances of the $π$-finite quantization procedure considered herein and the geometric quantization of a symplectic vector space.
