The generic tensor-rank conjecture for multipartite quantum states

Let H=\mathbbmCd1\mathbbmCdn\mathcal{H}=\mathbbm{C}^{d_1}\otimes\cdots\otimes\mathbbm{C}^{d_n} be the Hilbert space of multipartite quantum states. The expected tensor rank is the smallest kk for which the kk-secant variety fills the ambient projective space; equivalently, it is the ceiling of the ambient dimension divided by the dimension of a general rank-one parameter count. Generic tensor-rank conjecture. For a general multipartite quantum state, the generic tensor rank is

rkgen=i=1ndii=1n(di1)+1,\operatorname{rk}_{\text{gen}}=\left\lceil\frac{\prod_{i=1}^n d_i}{\sum_{i=1}^n(d_i-1)+1}\right\rceil,

except for \mathbbmC4×4×3\mathbbm{C}^{4\times4\times3}, \mathbbmC(2i+1)×(2i+1)×3\mathbbm{C}^{(2i+1)\times(2i+1)\times3}, and \mathbbmC(i+2)×(i+2)×2×2\mathbbm{C}^{(i+2)\times(i+2)\times2\times2} with i\mathbbmZ+i\in\mathbbm{Z}^+, where it is the expected rank plus one. This is the expected-dimension prediction for secant varieties of Segre varieties. The supplied text records the exceptional cases but does not state whether the conjecture has been resolved.

Sources & referencesView supporting material

Primary source

Masoud Gharahi, “Classifying Entanglement by Algebraic Geometry”, arXiv:2408.12265 (2025).

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.