On the Hikami-Inoue conjecture (Q1985993)

Given a \(3\)-manifold \(M\) with boundary consisting of a non-empty union of tori, a well-known strategy, due to Thurston, to assure that the interior of \(M\) admits a hyperbolic structure is to find an ideal triangulation which can be realised by hyperbolic ideal tetrahedra in such a way that the identifications induce a global hyperbolic structure with finite volume. The problem can be restated in algebraic terms: the existence of a hyperbolic structure for \(M\) corresponds to the existence of a solution of a set of algebraic equations where the variables are the moduli of the ideal tetrahedra of the triangulation. The same strategy can be adapted to find, more generally, representations of the fundamental group of \(M\) into \(\mathrm{SL}(n,{\mathbb C})\) which are parabolic on the peripheral groups, via the so called \textit{Ptolemy coordinates}. Note that the existence of a hyperbolic structure for \(M\) is equivalent to the existence of a discrete and faithful representation of \(\pi_1(M)\) into \(\mathrm{PSL}(2,{\mathbb C})\). If \(M\) is the complement of a knot \(K\) in the \(3\)-sphere, it is possible to define a decomposition into ideal octahedra of the manifold \(M\) with two points removed from a given a diagram presentation of \(K\): a neighbourhood of each crossing in the diagram can be seen as an ideal octahedron with four edges identified in two pairs, two vertices on \(K\), and two vertices on each of the two points removed. On its turn, such a decomposition gives rise to a decomposition into ideal tetrahedra (five for each octahedron). In [Algebr. Geom. Topol. 15, No. 4, 2175--2194 (2015; Zbl 1351.57009)], \textit{K. Hikami} and \textit{R. Inoue} considered the aforementioned triangulations of hyperbolic knot complements derived from planar diagrams of braid closures. For each braid, they provided equations whose solutions give rise to boundary parabolic represenations of the knot group into \(\mathrm{SL}(2,{\mathbb C})\), and conjectured that there exists a solution which is geometric, i.e. discrete and faithful. The authors show that Hikami and Inoue's conjecture holds if and only if the length of the braid is odd. In particular, by Alexander's and Markov's theorems, for every \(K\) it is always possible to find a braid whose closure is \(K\) and such that the associated equations admit a geometric solution.