Tannakian obstruction conjecture for braided fusion categories

Let GG be a finite group, let Rep(G)\mathbf{Rep}(G) denote its representation category, and let B\mathcal B be a braided fusion category with obstruction class O4(B)O_4(\mathcal B). Write \Obs(Rep(G))\Obs(\mathbf{Rep}(G)) for the obstruction group, \Witt\Witt for the Witt group, and \Witt(Rep(G))\Witt(\mathbf{Rep}(G)) for the Witt group relative to Rep(G)\mathbf{Rep}(G). Then O4(B)O_4(\mathcal B) induces an isomorphism

\Obs(Rep(G))coker(\Witt\Witt(Rep(G)))H4(BG;\mathbbmk×).\Obs(\mathbf{Rep}(G)) \cong \operatorname{coker}\bigl(\Witt \longrightarrow \Witt(\mathbf{Rep}(G))\bigr) \cong \operatorname{H}^4(\mathrm B G; \mathbbm{k}^\times).

Tannakian obstruction conjecture. The displayed isomorphism holds. This would complete the obstruction theory for Rep(G)\mathbf{Rep}(G) by identifying every class in H4(BG;\mathbbmk×)\operatorname{H}^4(\mathrm B G;\mathbbm{k}^\times) with the obstruction of a braided fusion category. The text sketches a construction using a finite-group extension that trivializes the pulled-back class, but the conjecture is presented subject to the assumptions made there.

Sources & referencesView supporting material

Primary source

Theo Johnson-Freyd and David Reutter, “Minimal nondegenerate extensions”, arXiv:2105.15167 (2023).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.