Computing Permanents and Counting Hamiltonian Cycles Faster

We show that the permanent of an $n\times n$ matrix of $\operatorname{poly}(n)$-bit integers and the number of Hamiltonian cycles of an $n$-vertex graph can both be computed in time $2^{n-Ω(\sqrt{n})}$, improving an earlier algorithm of Björklund, Kaski, and Williams (Algorithmica 2019) that runs in time $2^{n - Ω\left(\sqrt{n/\log \log n}\right)}$. A key tool of our approach is to design a data structure that supports fast "$r$-order evaluation" of permanent and Hamiltonian cycles, which cooperates with the new approach on multivariate multipoint evaluation by Bhargava, Ghosh, Guo, Kumar, and Umans (FOCS 2022).