Composite quantum Chernoff conjecture for finitely many hypotheses

At least 5 years old · documented by

Let ϱ\varrho and σ1,…,σr\sigma_1,\ldots,\sigma_r be quantum states, and define the sequences

ϱ⃗:={ϱ⊗n}n∈N,σ⃗i:={σi⊗n}n∈N(i∈[r]).\vec \varrho:=\{\varrho^{\otimes n}\}_{n\in\mathbb{N}},\qquad \vec \sigma_i:=\{\sigma_i^{\otimes n}\}_{n\in\mathbb{N}}\quad (i\in[r]).

For the optimal worst-case composite error probability, write pe(ϱ⃗,{σ⃗1,…,σ⃗r})p_e(\vec \varrho,\{\vec\sigma_1,\ldots,\vec\sigma_r\}) for the asymptotic error exponent when its limit exists.

Composite quantum Chernoff conjecture. The limit pe(ϱ⃗,{σ⃗1,…,σ⃗r})p_e(\vec \varrho,\{\vec\sigma_1,\ldots,\vec\sigma_r\}) exists, and

pe(ϱ⃗,{σ⃗1,…,σ⃗r})=max⁡ipe(ϱ⃗,σ⃗i).p_e(\vec \varrho,\{\vec\sigma_1,\ldots,\vec\sigma_r\})=\max_i p_e(\vec \varrho,\vec\sigma_i).

Equivalently, the composite exponent is determined by the least favorable individual hypothesis, or by the minimum of the corresponding Chernoff divergences. The conjecture is open even when r=2r=2, with one state in the first hypothesis and two states in the second.

References

Primary source

Zsombor Szilágyi, “On a conjecture regarding quantum hypothesis testing”, arXiv:2011.03342 (2020).

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.