Estimation of low rank density matrices: bounds in Schatten norms and other distances

Publication date 21 September 2016

Abstract: Let

{m a t h c a l S}_{m}

be the set of all

m i m e s m

density matrices (Hermitian positively semi-definite matrices of unit trace). Consider a problem of estimation of an unknown density matrix

h o i n {m a t h c a l S}_{m}

based on outcomes of

n

measurements of observables

X_{1}, d o t s, X_{n} i n {m a t h b b H}_{m}

(

{m a t h b b H}_{m}

being the space of

m i m e s m

Hermitian matrices) for a quantum system identically prepared

n

times in state

h o .

Outcomes

Y_{1}, d o t s, Y_{n}

of such measurements could be described by a trace regression model in which

{m a t h b b E}_{h o} (Y_{j} | X_{j}) = m t r (h o X_{j}), j = 1, d o t s, n .

The design variables

X_{1}, d o t s, X_{n}

are often sampled at random from the uniform distribution in an orthonormal basis

E_{1}, d o t s, E_{m^{2}}

of

{m a t h b b H}_{m}

(such as Pauli basis). The goal is to estimate the unknown density matrix

h o

based on the data

(X_{1}, Y_{1}), d o t s, (X_{n}, Y_{n}) .

Let

hat Z:=frac{m^2}{n}sum_{j=1}^n Y_j X_j

and let

c h e c k h o

be the projection of

h a t Z

onto the convex set

{m a t h c a l S}_{m}

of density matrices. It is shown that for estimator

c h e c k h o

the minimax lower bounds in classes of low rank density matrices (established earlier) are attained up logarithmic factors for all Schatten

p

-norm distances,

p i n [1, i n f t y]

and for Bures version of quantum Hellinger distance. Moreover, for a slightly modified version of estimator

c h e c k h o

the same property holds also for quantum relative entropy (Kullback-Leibler) distance between density matrices.

Recommendations

Cites work

Cited in

(8)

This page was built for publication: Estimation of low rank density matrices: bounds in Schatten norms and other distances