Skip to main content

Improvements of operator reverse AM-GM inequality involving positive linear maps

Abstract

In this paper, we shall present some reverse arithmetic-geometric mean operator inequalities for unital positive linear maps. These inequalities improve some corresponding results due to Xue (J. Inequal. Appl. 2017:283, 2017).

Introduction

Let m, \({m^{\prime}} \), \(m_{1}^{\prime}\), \(m_{2}^{\prime}\), \(m_{3}^{\prime}\), M, \({M^{\prime}}\), \(M_{1}^{\prime}\), \(M_{2}^{\prime}\) and \(M_{3}^{\prime}\) be scalars, I be the identity operator and the other capital letters be used to represent general elements of the \(C^{*}\)-algebra \(\mathcal{B}(\mathcal {H})\) of all bounded linear operators acting on a Hilbert space\((\mathcal{H},\langle\cdot,\cdot\rangle)\). The operator norm is denoted by \(\|\cdot\|\). An operator A is said to be positive if \(\langle Ax,x\rangle\geq0\) for all \(x \in\mathcal{H}\) and we write it as \(A\geq0\), it is said to be strictly positive if \(\langle Ax,x\rangle>0\) for all \(x \in {\mathcal{H}\setminus\{0\}}\) and we write it as \(A>0\). A linear map Φ is positive if \(\varPhi(A)\geq0\) whenever \(A\geq0\). It is said to be unital if \(\varPhi(I)=I\). For \(A, B>0\) the μ-weighted arithmetic mean and μ-weighted geometric mean of A and B are defined, respectively, by

$$ A\bigtriangledown_{\mu}B=(1-\mu)A+\mu B, \qquad A \sharp_{\mu}B = A^{1/2}\bigl(A^{-1/2}B A^{-1/2} \bigr)^{\mu}A^{1/2}, $$

where \(\mu\in[0,1]\), when \(\mu=1/2\), we write \(A\nabla B\) and \(A\sharp B \) for brevity for \(A\nabla_{1/2}B\) and \(A \sharp_{1/2}B\), respectively.

For \(0< m\leq A,B\leq M\), Tominaga [2] proved that the following operator reverse AM-GM inequality holds:

$$\begin{aligned} \frac{A+B}{2}\leq S(h) A\sharp B, \end{aligned}$$
(1.1)

where \(S(h)=\frac{h^{\frac{1}{h-1}}}{e\log h^{\frac{1}{h-1}}}\) is called Specht’s ratio with \(h =\frac{M}{m}\).

The inequality (1.1) can be regarded as a counterpart of the following AM-GM inequality:

$$\begin{aligned} \frac{A+B}{2}\geq A\sharp B. \end{aligned}$$
(1.2)

Lin [3, (3.3)] observed that

$$\begin{aligned} S(h)\leq K(h)\leq S^{2}(h) \quad(h\geq1), \end{aligned}$$
(1.3)

where \(K(h)=\frac{(h+1)^{2}}{4h} \) and \(h=\frac{M}{m}\). The constant \(K(t,2)=\frac{(t+1)^{2}}{4t}\) (\(t>0\)) is called the Kantorovich constant, which is simply represented by \(K(t)\) satisfying the following properties:

  • \(K(1,2)=1\),

  • \(K(t,2)= K(\frac{1}{t} ,2)\geq1\) (\(t>0\)),

  • \(K(t,2)\) is monotone increasing on \([1,\infty)\) and monotone decreasing on \((0,1]\).

By inequalities (1.1) and (1.3), we have

$$\begin{aligned} \frac{A+B}{2}\leq K(h) A\sharp B. \end{aligned}$$
(1.4)

Because Φ is order preserving, (1.4) implies that

$$\begin{aligned} \varPhi \biggl(\frac{A+B}{2} \biggr)\leq K(h)\varPhi(A\sharp B). \end{aligned}$$
(1.5)

For a positive linear map Φ and \(A,B\geq0\). Ando [4] has proved the following inequality:

$$\begin{aligned} \varPhi(A\sharp B)\leq\bigl(\varPhi(A)\sharp\varPhi(B)\bigr). \end{aligned}$$
(1.6)

Then, by (1.5) and (1.6), we have

$$\begin{aligned} \varPhi \biggl(\frac{A+B}{2} \biggr)\leq K(h) \bigl(\varPhi(A)\sharp\varPhi(B)\bigr). \end{aligned}$$
(1.7)

The studies of squaring operator inequalities start with [3, 5] and continued by a number of authors [610]. Lin [3] revealed that inequalities (1.5) and (1.7) can be squared as follows:

$$\begin{aligned}& \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq K^{2}(h) \varPhi^{2}(A\sharp B), \end{aligned}$$
(1.8)
$$\begin{aligned}& \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq K^{2}(h) \bigl( \varPhi (A)\sharp\varPhi(B)\bigr)^{2}. \end{aligned}$$
(1.9)

Recently, Xue [1] proved that if \(\sqrt{\frac{M}{m}}\leq2.314 \), then the following refinement of the inequality(1.4) holds:

$$\begin{aligned} \biggl(\frac{A+B}{2} \biggr)\leq K^{\frac{1}{2}}(h) (A\sharp B). \end{aligned}$$
(1.10)

Inspired by Lin’s idea [3] , Xue [1] also proved that if \(0< m\leq A, B\leq M\) and \(\sqrt{\frac{M}{m}}\leq2.314 \), then

$$\begin{aligned}& \biggl(\frac{A+B}{2} \biggr)^{2}\leq K(h) (A\sharp B)^{2}, \end{aligned}$$
(1.11)
$$\begin{aligned}& \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq K(h) \varPhi^{2}(A\sharp B), \end{aligned}$$
(1.12)

and

$$\begin{aligned} &\varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq K(h) \bigl(\varPhi(A) \sharp\varPhi(B)\bigr)^{2}, \end{aligned}$$
(1.13)

inequalities (1.12) and (1.13) are refinements of the inequalities (1.8) and (1.9), respectively.

Moreover, she proved Lin’s conjecture [3] as follows:

$$\begin{aligned}& \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq S^{2}(h) \varPhi^{2}(A\sharp B), \end{aligned}$$
(1.14)
$$\begin{aligned}& \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq S^{2}(h) \bigl( \varPhi (A)\sharp\varPhi(B)\bigr)^{2}. \end{aligned}$$
(1.15)

Recently, Ali et al. obtained more refinements of the results presented by Xue [1] by using the relation (1.2), for comprehensive study, the reader is referred to [11]. In this article, in Sect. 2, we shall refine the inequalities (1.10)–(1.15), when \(\sqrt{\frac{M}{m}}\leq 2.314\), with the help of the Kantorovich constant.

Main results

We begin this section with the following lemmas.

Lemma 2.1

([12])

Let\(A,B >0\). Then the following norm inequality holds:

$$\begin{aligned} \Vert AB \Vert \leq\frac{1}{4} \Vert A+B \Vert ^{2}. \end{aligned}$$
(2.1)

Remark 2.2

Lemma 2.1 is proved by Bhatia and Kittaneh in [12] for the finite dimensional case. However, all technical results used to prove this result for operator norm are also true for the infinite dimensional case. Here also, we mention that if \(A,B\) are compact operators, then a stronger result can be found in [13].

Lemma 2.3

([14])

Let\(A>0\). Then, for every positive unital linear mapΦ,

$$\begin{aligned} \varPhi^{-1}(A)\leq\varPhi\bigl(A^{-1}\bigr). \end{aligned}$$
(2.2)

Lemma 2.4

([15])

Suppose that two operatorsA, Band positive real numbersm, \({m^{\prime}}\), M, \({M^{\prime}}\)satisfy either of the following conditions:

  1. (1)

    \(0< m\leq A\leq m^{\prime}< M^{\prime}\leq B\leq M\),

  2. (2)

    \(0< m\leq B\leq m^{\prime}< M^{\prime}\leq A\leq M\).

Then

$$\begin{aligned} K^{r}\bigl(h^{\prime}\bigr) \bigl( A^{-1} \sharp_{\mu} B^{-1} \bigr)\leq A^{-1}\nabla _{\mu}B^{-1}, \end{aligned}$$
(2.3)

for all\(\mu\in[0,1]\), \(r=\min[\mu, 1-\mu]\), \(h=\frac{M}{m}\)and\(h^{\prime}=\frac{{M^{\prime}}}{m^{\prime}}\).

Now, we prove the first main result in the following theorem.

Theorem 2.5

Let\(0< m\leq M\)and\(\sqrt{\frac{M}{m}}\leq2.314 \), we have

  1. (1)

    If\(0< m\leq A\leq m_{1}^{\prime}< M_{1}^{\prime}\leq B\leq\frac{M+m}{2}\), then

    $$\begin{aligned} \biggl(\frac{A+B}{2} \biggr)^{2}\leq\frac{K(h)}{K(h_{1}^{\prime})}(A\sharp B)^{2}, \end{aligned}$$
    (2.4)

    where\(K(h)= \frac{(h+1)^{2}}{4h}\), \(K(h_{1}^{\prime})=\frac{(h_{1}^{\prime}+1)^{2}}{4h_{1}^{\prime}}\), \(h = \frac{M}{m}\)and\(h_{1}^{\prime}=\frac {M_{1}^{\prime}}{m_{1}^{\prime}}\).

  2. (2)

    If\(0<\frac{M+m}{2}\leq A\leq m_{2}^{\prime}<M_{2}^{\prime}\leq B\leq M\), then

    $$\begin{aligned} \biggl(\frac{A+B}{2} \biggr)^{2}\leq\frac{K(h)}{K(h_{2}^{\prime})}(A\sharp B)^{2}, \end{aligned}$$
    (2.5)

    where\(K(h)= \frac{(h+1)^{2}}{4h}\), \(K(h_{2}^{\prime})= \frac{(h_{2}^{\prime}+1)^{2}}{4h^{\prime}_{2}}\), \(h = \frac{M}{m}\)and\(h_{2}^{\prime}= \frac {M_{2}^{\prime}}{m_{2}^{\prime}}\).

  3. (3)

    If\(0< m\leq A\leq m_{3}^{\prime}<\frac{M+m}{2}\leq B\leq M\), then

    $$\begin{aligned} \biggl(\frac{A+B}{2} \biggr)^{2}\leq\frac{K(h)}{K(h_{3}^{\prime})}(A\sharp B)^{2} , \end{aligned}$$
    (2.6)

    where\(K(h)= \frac{(h+1)^{2}}{4h}\), \(K(h_{3}^{\prime})= \frac{(h_{3}^{\prime}+1)^{2}}{4h_{3}^{\prime}}\), \(h = \frac{M}{m}\)and\(h_{3}^{\prime}= \frac {M+m}{2m_{3}^{\prime}}\).

  4. (4)

    If\(0< m\leq A\leq\frac{M+m}{2}<M_{3}^{\prime}\leq B\leq M\), then

    $$\begin{aligned} \biggl(\frac{A+B}{2} \biggr)^{2}\leq\frac{K(h)}{K(h_{4}^{\prime})}(A\sharp B)^{2} , \end{aligned}$$
    (2.7)

    where\(K(h)= \frac{(h+1)^{2}}{4h}\), \(K(h_{4}^{\prime})= \frac{(h_{4}^{\prime}+1)^{2}}{4h^{\prime}_{4}}\), \(h = \frac{M}{m}\)and\(h_{4}^{\prime}= \frac {2M_{3}^{\prime}}{M+m}\).

Proof

The operator inequality (2.4) is equivalent to

$$\begin{aligned} \biggl\Vert \frac{A+B}{2}(A\sharp B)^{-1} \biggr\Vert \leq \frac{K^{\frac {1}{2}}(h)}{{K^{\frac{1}{2}}}(h_{1}^{\prime})}. \end{aligned}$$

If \(0< m\leq A\leq m_{1}^{\prime}< M_{1}^{\prime}\leq B\leq\frac{M+m}{2}\), we get

$$\begin{aligned} A+\frac{M+m}{2}m A^{-1}\leq\frac{M+m}{2}+m \end{aligned}$$
(2.8)

and

$$\begin{aligned} B+\frac{M+m}{2}m B^{-1}\leq\frac{M+m}{2}+m. \end{aligned}$$
(2.9)

Compute

$$\begin{aligned} & \biggl\Vert \frac{A+B}{2}\frac{M+m}{2}.mK^{\frac{1}{2}} \bigl(h_{1}^{\prime}\bigr) (A\sharp B)^{-1} \biggr\Vert \\ &\quad\leq\frac{1}{4} \biggl\Vert \frac{A+B}{2}+\frac{M+m}{2}.m K^{\frac {1}{2}}\bigl(h_{1}^{\prime}\bigr) (A\sharp B)^{-1} \biggr\Vert ^{2}\quad\bigl(\text{by (2.1)}\bigr) \\ &\quad=\frac{1}{4} \biggl\Vert \frac{A+B}{2}+\frac{M+m}{2}.m K^{\frac {1}{2}}\bigl(h_{1}^{\prime}\bigr) \bigl(A^{-1} \sharp B^{-1}\bigr) \biggr\Vert ^{2} \\ &\quad\leq\frac{1}{4} \biggl\Vert \frac{A+B}{2}+\frac{M+m}{2}.m \frac {A^{-1}+B^{-1}}{2} \biggr\Vert ^{2} \quad\bigl(\text{by (2.3)}\bigr) \\ &\quad\leq\frac{1}{4} \biggl(\frac{M+m}{2}+m \biggr)^{2}\quad \bigl(\text{by (2.8), (2.9)}\bigr). \end{aligned}$$

That is,

$$\begin{aligned} \biggl\Vert \frac{A+B}{2}(A\sharp B)^{-1} \biggr\Vert \leq \frac{(\frac {M+m}{2}+m)^{2}}{4\frac{M+m}{2}.mK^{\frac{1}{2}}(h_{1}^{\prime})}. \end{aligned}$$

Since \(1\leq\sqrt{\frac{M}{m}}\leq2.314\), it follows that

$$\begin{aligned} \biggl(\sqrt{\frac{M}{m}}-1 \biggr)^{2} \biggl[ \biggl(\sqrt{ \frac{M}{m}} \biggr)^{3}-\frac{2M}{m}+\sqrt{ \frac{M}{m}}-4 \biggr]\leq0. \end{aligned}$$
(2.10)

It is easy to see that \(\frac{ (\frac{M+m}{2}+m )^{2}}{4\frac{M+m}{2}.m}\leq\frac {M+m}{2\sqrt{M m}}\) is equivalent to (2.10).

Thus

$$\begin{aligned} \biggl\Vert \frac{A+B}{2}(A\sharp B)^{-1} \biggr\Vert \leq\frac{M+m}{2\sqrt {Mm}K^{\frac{1}{2}}(h_{1}^{\prime})}=\frac{K^{\frac{1}{2}}(h)}{K^{\frac {1}{2}}(h_{1}^{\prime})}. \end{aligned}$$

If \(0<\frac{M+m}{2}\leq A\leq m_{2}^{\prime}<M_{2}^{\prime}\leq B\leq M\), we get

$$\begin{aligned}& A+\frac{M+m}{2}MA^{-1}\leq\frac{M+m}{2}+M, \end{aligned}$$
(2.11)
$$\begin{aligned}& B+\frac{M+m}{2}M B^{-1}\leq\frac{M+m}{2}+M. \end{aligned}$$
(2.12)

Similarly, we have

$$\begin{aligned} \biggl\Vert \frac{A+B}{2}(A\sharp B)^{-1} \biggr\Vert \leq \frac{(\frac {M+m}{2}+M)^{2}}{4\frac{M+m}{2}.M K^{\frac{1}{2}}(h_{2}^{\prime})}. \end{aligned}$$
(2.13)

Since \(\frac{(\frac{M+m}{2}+M)^{2}}{4\frac{M+m}{2}.M}\leq\frac{(\frac {M+m}{2}+m)^{2}}{4\frac{M+m}{2}.m}\leq\frac{M+m}{2\sqrt{M m}}\), so (2.13) becomes

$$\begin{aligned} \biggl\Vert \frac{A+B}{2}(A\sharp B)^{-1} \biggr\Vert \leq\frac{M+m}{2\sqrt {Mm}K^{\frac{1}{2}}(h_{2}^{\prime})}=\frac{K^{\frac{1}{2}}(h)}{K^{\frac {1}{2}}(h_{2}^{\prime})}. \end{aligned}$$

If \(0< m\leq A\leq m_{3}^{\prime}<\frac{M+m}{2}\leq B\leq M\), then we compute

$$\begin{aligned} & \biggl\Vert \frac{A+B}{2}\frac{M+m}{2}.K^{\frac{1}{2}} \bigl(h_{3}^{\prime}\bigr)\sqrt{m M} (A\sharp B )^{-1} \biggr\Vert \\ &\quad\leq\frac{1}{4} \biggl\Vert \frac{A+B}{2}+\frac{M+m}{2}.K^{\frac{1}{2}} \bigl(h_{3}^{\prime}\bigr)\sqrt{mM} (A\sharp B )^{-1} \biggr\Vert ^{2} \quad\bigl(\text{by(2.1)}\bigr) \\ &\quad=\frac{1}{4} \biggl\Vert \frac{A+B}{2}+\frac{M+m}{2}.K^{\frac{1}{2}} \bigl(h_{3}^{\prime}\bigr)\sqrt{mM} \bigl(A^{-1}\sharp B^{-1} \bigr) \biggr\Vert ^{2} \\ &\quad=\frac{1}{4} \biggl\Vert \frac{A+B}{2}+\frac{M+m}{2}. K^{\frac {1}{2}}\bigl(h_{3}^{\prime}\bigr) \bigl(mA^{-1} \sharp MB^{-1} \bigr) \biggr\Vert ^{2} \\ &\quad\leq\frac{1}{4} \biggl\Vert \frac{A+B}{2}+\frac{M+m}{2}. \frac {mA^{-1}+MB^{-1}}{2} \biggr\Vert ^{2} \quad\bigl(\text{by (2.3)}\bigr) \\ &\quad\leq\frac{1}{4}(M+m)^{2} \quad\bigl(\text{by (2.8), (2.12)}\bigr), \end{aligned}$$
(2.14)

so we have

$$\begin{aligned} \biggl\Vert \frac{A+B}{2} (A\sharp B )^{-1} \biggr\Vert \leq \frac{(M+m)^{2}}{4\frac{M+m}{2}\sqrt{Mm} K^{\frac{1}{2}}(h_{3}^{\prime})}= \frac{M+m}{{2\sqrt{Mm}}K^{\frac{1}{2}}(h_{3}^{\prime})} = \frac{K^{\frac {1}{2}}(h)}{K^{\frac{1}{2}}(h_{3}^{\prime})}. \end{aligned}$$

If \(0< m\leq A\leq\frac{M+m}{2}<M_{3}^{\prime}\leq B\leq M\), similarly, by (2.1), (2.3), (2.8) and (2.12), we have

$$\begin{aligned} \biggl\Vert \frac{A+B}{2} (A\sharp B )^{-1} \biggr\Vert \leq \frac {M+m}{{2\sqrt{Mm}}K^{\frac{1}{2}}(h_{4}^{\prime})} = \frac{K^{\frac {1}{2}}(h)}{K^{\frac{1}{2}}(h_{4}^{\prime})}. \end{aligned}$$

This completes the proof. □

Remark 2.6

Because \(\frac{K(h)}{K(h_{1}^{\prime})}< K(h)\), \(\frac {K(h)}{K(h_{2}^{\prime})}< K(h)\), \(\frac{K(h)}{K(h_{3}^{\prime})}< K(h)\) and \(\frac{K(h)}{K(h_{4}^{\prime})}< K(h)\), so Theorem 2.5 is a refinement of the inequality (1.11).

Remark 2.7

Since \(t^{p}\) is operator monotone function for \(0\leq p\leq1\), so, by taking power \(\frac{1}{2}\) both sides of (2.4), (2.5), (2.6) and (2.7), respectively, we can easily get a refinement of the inequality (1.10) for the condition \(\sqrt{\frac{M}{m}}\leq2.314 \).

Theorem 2.8

Let\(0< m\leq M\)and\(\sqrt{\frac{M}{m}}\leq 2.314\), we have

  1. (1)

    If\(0< m\leq A\leq m_{1}^{\prime}< M_{1}^{\prime}\leq B\leq\frac{M+m}{2}\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{K(h)}{K(h_{1}^{\prime})}\varPhi ^{2}(A\sharp B), \end{aligned}$$
    (2.15)

    where\(K(h)= \frac{(h+1)^{2}}{4h}\), \(K(h_{1}^{\prime})=\frac{(h_{1}^{\prime}+1)^{2}}{4h_{1}^{\prime}}\), \(h = \frac{M}{m}\)and\(h_{1}^{\prime}=\frac {M_{1}^{\prime}}{m_{1}^{\prime}}\).

  2. (2)

    If\(0<\frac{M+m}{2}\leq A\leq m_{2}^{\prime}<M_{2}^{\prime}\leq B\leq M\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{K(h)}{K(h_{2}^{\prime})}\varPhi ^{2}(A\sharp B), \end{aligned}$$
    (2.16)

    where\(K(h)= \frac{(h+1)^{2}}{4h}\), \(K(h_{2}^{\prime})= \frac{(h_{2}^{\prime}+1)^{2}}{4h^{\prime}_{2}}\), \(h = \frac{M}{m}\)and\(h_{2}^{\prime}= \frac {M_{2}^{\prime}}{m_{2}^{\prime}}\).

  3. (3)

    If\(0< m\leq A\leq m_{3}^{\prime}<\frac{M+m}{2}\leq B\leq M\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{K(h)}{K(h_{3}^{\prime})}\varPhi ^{2}(A\sharp B), \end{aligned}$$
    (2.17)

    where\(K(h)= \frac{(h+1)^{2}}{4h}\), \(K(h_{3}^{\prime})= \frac{(h_{3}^{\prime}+1)^{2}}{4h_{3}^{\prime}}\), \(h = \frac{M}{m}\)and\(h_{3}^{\prime}= \frac {M+m}{2m_{3}^{\prime}}\).

  4. (4)

    If\(0< m\leq A\leq\frac{M+m}{2}<M_{3}^{\prime}\leq B\leq M\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{K(h)}{K(h_{4}^{\prime})}\varPhi ^{2}(A\sharp B), \end{aligned}$$
    (2.18)

    where\(K(h)= \frac{(h+1)^{2}}{4h}\), \(K(h_{4}^{\prime})= \frac{(h_{4}^{\prime}+1)^{2}}{4h^{\prime}_{4}}\), \(h = \frac{M}{m}\)and\(h_{4}^{\prime}= \frac {2M_{3}^{\prime}}{M+m}\).

Proof

Inequality (2.15) is equivalent to

$$\begin{aligned} \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)\varPhi^{-1}(A \sharp B) \biggr\Vert \leq \frac{K^{\frac{1}{2}}(h)}{{K^{\frac{1}{2}}}(h_{1}^{\prime})}. \end{aligned}$$

If \(0< m\leq A\leq m_{1}^{\prime}< M_{1}^{\prime}\leq B\leq\frac{M+m}{2}\), then we compute

$$\begin{aligned} & \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)\frac{M+m}{2}.mK^{\frac {1}{2}} \bigl(h_{1}^{\prime}\bigr)\varPhi^{-1} (A\sharp B ) \biggr\Vert \\ &\quad\leq\frac{1}{4} \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)+ \frac{M+m}{2}.m K^{\frac{1}{2}}\bigl(h_{1}^{\prime}\bigr) \varPhi^{-1} (A\sharp B ) \biggr\Vert ^{2} \quad\bigl(\text{by (2.1)}\bigr) \\ &\quad\leq\frac{1}{4} \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)+ \frac{M+m}{2}.m K^{\frac{1}{2}}\bigl(h_{1}^{\prime}\bigr)\varPhi \bigl((A\sharp B)^{-1} \bigr) \biggr\Vert ^{2} \quad\bigl(\text{by (2.2)}\bigr) \\ &\quad=\frac{1}{4} \biggl\Vert \varPhi \biggl(\frac{A+B}{2}+ \frac{M+m}{2}.m K^{\frac {1}{2}}\bigl(h_{1}^{\prime}\bigr) \bigl(A^{-1}\sharp B^{-1}\bigr) \biggr) \biggr\Vert ^{2} \\ &\quad\leq\frac{1}{4} \biggl\Vert \varPhi \biggl(\frac{A+B}{2}+ \frac{M+m}{2}.m \frac {A^{-1}+B^{-1}}{2} \biggr) \biggr\Vert ^{2} \quad\bigl(\text{by (2.3)}\bigr) \\ &\quad\leq\frac{1}{4} \biggl(\frac{M+m}{2}+m \biggr)^{2}\quad \bigl(\text{by (2.8), (2.9)}\bigr), \end{aligned}$$
(2.19)

that is,

$$\begin{aligned} \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)\varPhi^{-1}(A\sharp B) \biggr\Vert \leq \frac{ (\frac{M+m}{2}+m )^{2}}{4\frac{M+m}{2}.mK^{\frac {1}{2}}(h_{1}^{\prime})}. \end{aligned}$$
(2.20)

By \(1\leq\sqrt{\frac{M}{m}}\leq2.314\) and (2.10), we have

$$\begin{aligned} \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)\varPhi^{-1}(A \sharp B) \biggr\Vert \leq \frac{M+m}{2\sqrt{Mm}K^{\frac{1}{2}}(h_{1}^{\prime})}=\frac{K^{\frac {1}{2}}(h)}{{K^{\frac{1}{2}}}(h_{1}^{\prime})}. \end{aligned}$$

If \(0<\frac{M+m}{2}\leq A\leq m_{2}^{\prime}<M_{2}^{\prime}\leq B\leq M\), similarly, by (2.1), (2.2), (2.3), (2.11), (2.12), \(\frac{(\frac{M+m}{2}+M)^{2}}{4\frac{M+m}{2}M}\leq\frac{(\frac {M+m}{2}+m)^{2}}{4\frac{M+m}{2}m}\) and (2.10), we obtain

$$\begin{aligned} \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)\varPhi^{-1}(A \sharp B) \biggr\Vert \leq\frac{M+m}{2\sqrt{Mm}K^{\frac{1}{2}}(h_{2}^{\prime})}=\frac{K^{\frac {1}{2}}(h)}{{K^{\frac{1}{2}}}(h_{2}^{\prime})}. \end{aligned}$$

If \(0< m\leq A\leq m_{3}^{\prime}<\frac{M+m}{2}\leq B\leq M\), then we compute

$$\begin{aligned} & \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)\frac{M+m}{2}.K^{\frac {1}{2}} \bigl(h_{3}^{\prime}\bigr)\sqrt{mM}\varPhi^{-1}(A\sharp B) \biggr\Vert \\ &\quad\leq\frac{1}{4} \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)+ \frac {M+m}{2}.K^{\frac{1}{2}}\bigl(h_{3}^{\prime}\bigr) \sqrt{mM}\varPhi^{-1}(A\sharp B) \biggr\Vert ^{2} \quad\bigl(\text{by (2.1)} \bigr) \\ &\quad\leq\frac{1}{4} \biggl\Vert \varPhi \biggl(\frac {A+B}{2} \biggr)+ \frac{M+m}{2}.K^{\frac{1}{2}}\bigl(h_{3}^{\prime}\bigr) \sqrt{mM}\varPhi \bigl((A\sharp B)^{-1}\bigr) \biggr\Vert ^{2} \quad\bigl(\text{by (2.2)}\bigr) \\ &\quad=\frac{1}{4} \biggl\Vert \varPhi \biggl(\frac{A+B}{2}+ \frac{M+m}{2}.K^{\frac{1}{2}}\bigl(h_{3}^{\prime}\bigr)\sqrt {mM}\bigl(A^{-1}\sharp B^{-1}\bigr) \biggr) \biggr\Vert ^{2} \\ &\quad=\frac {1}{4} \biggl\Vert \varPhi \biggl(\frac{A+B}{2}+ \frac{M+m}{2}.K^{\frac {1}{2}}\bigl(h_{3}^{\prime}\bigr) \bigl(mA^{-1}\sharp MB^{-1}\bigr) \biggr) \biggr\Vert ^{2} \\ &\quad\leq\frac{1}{4} \biggl\Vert \varPhi \biggl(\frac{A+B}{2}+ \frac{M+m}{2}.\frac {mA^{-1}+ MB^{-1}}{2} \biggr) \biggr\Vert ^{2} \quad\bigl(\text{by (2.3)}\bigr) \\ &\quad\leq\frac {(M+m)^{2}}{4} \quad\bigl(\text{by (2.8), (2.12)}\bigr), \end{aligned}$$
(2.21)

that is, we have

$$\begin{aligned} \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)\varPhi^{-1}(A\sharp B) \biggr\Vert \leq \frac{(M+m)^{2}}{4\frac{M+m}{2}\sqrt{M m}K^{\frac{1}{2}}(h_{3}^{\prime})}=\frac {M+m}{2\sqrt{Mm}K^{\frac{1}{2}}(h_{3}^{\prime})} =\frac{K^{\frac{1}{2}}(h)}{K^{\frac{1}{2}}(h_{3}^{\prime})}. \end{aligned}$$

If \(0< m\leq A\leq\frac{M+m}{2}<M_{3}^{\prime}\leq B\leq M\), similarly, by (2.1), (2.2), (2.3), (2.8) and (2.12), we have

$$\begin{aligned} \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)\varPhi^{-1}(A\sharp B) \biggr\Vert \leq \frac{M+m}{2\sqrt{Mm}K^{\frac{1}{2}}(h_{4}^{\prime})} =\frac{K^{\frac{1}{2}}(h)}{K^{\frac{1}{2}}(h_{4}^{\prime})}. \end{aligned}$$

It completes the proof. □

Remark 2.9

Obviously, Theorem 2.8 is a refinement of (1.12).

Theorem 2.10

Let\(0< m\leq M\)and\(\sqrt{\frac{M}{m}}\leq 2.314\), we have

  1. (1)

    If\(0< m\leq A\leq m_{1}^{\prime}< M_{1}^{\prime}\leq B\leq\frac{M+m}{2}\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{K(h)}{K(h_{1}^{\prime})}\bigl( \varPhi (A)\sharp\varPhi(B)\bigr)^{2}, \end{aligned}$$
    (2.22)

    where\(K(h)= \frac{(h+1)^{2}}{4h}\), \(K(h_{1}^{\prime})= \frac{(h_{1}^{\prime}+1)^{2}}{4h_{1}^{\prime}}\), \(h = \frac{M}{m}\)and\(h_{1}^{\prime}= \frac {M_{1}^{\prime}}{m_{1}^{\prime}}\).

  2. (2)

    If\(0<\frac{M+m}{2}\leq A\leq m_{2}^{\prime}<M_{2}^{\prime}\leq B\leq M\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{K(h)}{K(h_{2}^{\prime})}\bigl( \varPhi (A)\sharp\varPhi(B)\bigr)^{2}, \end{aligned}$$
    (2.23)

    where\(K(h)= \frac{(h+1)^{2}}{4h}\), \(K(h_{2}^{\prime})= \frac{(h_{2}^{\prime}+1)^{2}}{4h^{\prime}_{2}}\), \(h = \frac{M}{m}\)and\(h_{2}^{\prime}= \frac {M_{2}^{\prime}}{m_{2}^{\prime}}\).

  3. (3)

    If\(0< m\leq A\leq m_{3}^{\prime}<\frac{M+m}{2}\leq B\leq M\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{K(h)}{K(h_{3}^{\prime})}\bigl( \varPhi (A)\sharp\varPhi(B)\bigr)^{2}, \end{aligned}$$
    (2.24)

    where\(K(h)= \frac{(h+1)^{2}}{4h}\), \(K(h_{3}^{\prime})= \frac{(h_{3}^{\prime}+1)^{2}}{4h_{3}^{\prime}}\), \(h = \frac{M}{m}\)and\(h_{3}^{\prime}= \frac {M+m}{2m_{3}^{\prime}}\).

  4. (4)

    If\(0< m\leq A\leq\frac{M+m}{2}<M_{3}^{\prime}\leq B\leq M\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{K(h)}{K(h_{4}^{\prime})}\bigl( \varPhi (A)\sharp\varPhi(B)\bigr)^{2}, \end{aligned}$$
    (2.25)

    where\(K(h)= \frac{(h+1)^{2}}{4h}\), \(K(h_{4}^{\prime})= \frac{(h_{4}^{\prime}+1)^{2}}{4h^{\prime}_{4}}\), \(h = \frac{M}{m}\)and\(h_{4}^{\prime}= \frac {2M_{3}^{\prime}}{M+m}\).

Proof

Inequality (2.22) is equivalent to

$$\begin{aligned} \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr) \bigl(\varPhi(A)\sharp \varPhi(B)\bigr)^{-1} \biggr\Vert \leq\frac{K^{\frac{1}{2}}(h)}{{K^{\frac{1}{2}}}(h_{1}^{\prime})}. \end{aligned}$$

If \(0< m\leq A\leq m_{1}^{\prime}< M_{1}^{\prime}\leq B\leq\frac{M+m}{2}\), then we compute

$$\begin{aligned} & \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)\frac{M+m}{2}.mK^{\frac {1}{2}} \bigl(h_{1}^{\prime}\bigr) \bigl(\varPhi(A)\sharp\varPhi(B) \bigr)^{-1} \biggr\Vert \\ &\quad\leq\frac{1}{4}\|\varPhi \biggl(\frac{A+B}{2} \biggr)+ \frac{M+m}{2}.m K^{\frac{1}{2}}\bigl(h_{1}^{\prime}\bigr) \bigl(\varPhi(A)\sharp\varPhi B\bigr))^{-1}\|^{2} \quad\bigl(\text{by (2.1)} \bigr) \\ &\quad\leq\frac{1}{4} \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)+ \frac{M+m}{2}.m K^{\frac{1}{2}}\bigl(h_{1}^{\prime}\bigr) \varPhi^{-1}(A\sharp B) \biggr\Vert ^{2}\quad \bigl(\text{by (1.6)}\bigr) \\ &\quad\leq\frac{1}{4} \biggl(\frac{M+m}{2}+m \biggr)^{2} \quad\bigl(\text{by (2.19)}\bigr), \end{aligned}$$

that is,

$$\begin{aligned} \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr) \bigl(\varPhi(A)\sharp\varPhi( B) \bigr)^{-1} \biggr\Vert \leq\frac{ (\frac{M+m}{2}+m )^{2}}{4\frac {M+m}{2}.mK^{\frac{1}{2}}(h_{1}^{\prime})}. \end{aligned}$$

By \(1\leq\sqrt{\frac{M}{m}}\leq2.314\) and (2.10), we have

$$\begin{aligned} \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)\varPhi \bigl((A)\sharp \varPhi(B) \bigr)^{-1} \biggr\Vert \leq\frac{M+m}{2\sqrt{Mm}K^{\frac{1}{2}}(h_{1}^{\prime})}= \frac {K^{\frac{1}{2}}(h)}{{K^{\frac{1}{2}}}(h_{1}^{\prime})}. \end{aligned}$$

Since \(\frac{(\frac{M+m}{2}+M)^{2}}{4\frac{M+m}{2}M}\leq\frac{(\frac {M+m}{2}+m)^{2}}{4\frac{M+m}{2}m}\), by 2nd case \(0<\frac{M+m}{2}\leq A\leq m_{2}^{\prime}<M_{2}^{\prime}\leq B\leq M\), we can easily obtain the inequality (2.23).

If \(0< m\leq A\leq m_{3}^{\prime}<\frac{M+m}{2}\leq B\leq M\), then we compute

$$\begin{aligned} & \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)\frac{M+m}{2}.K^{\frac {1}{2}} \bigl(h_{3}^{\prime}\bigr)\sqrt{mM}\bigl(\varPhi(A)\sharp\varPhi(B) \bigr)^{-1} \biggr\Vert \\ &\quad\leq\frac{1}{4} \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)+ \frac {M+m}{2}.K^{\frac{1}{2}}\bigl(h_{3}^{\prime}\bigr) \sqrt{mM}\bigl(\varPhi(A)\sharp\varPhi (B)\bigr)^{-1} \biggr\Vert ^{2} \quad\bigl(\text{by (2.1)}\bigr) \\ &\quad\leq\frac{1}{4} \biggl\Vert \varPhi \biggl(\frac{A+B}{2} \biggr)+ \frac {M+m}{2}.K^{\frac{1}{2}}\bigl(h_{3}^{\prime}\bigr) \sqrt{mM}\varPhi^{-1}(A\sharp B) \biggr\Vert ^{2} \quad\bigl(\text{by (1.6)} \bigr) \\ &\quad\leq\frac{1}{4}(M+m)^{2}\quad \bigl(\text{by (2.21)}\bigr), \end{aligned}$$

thus, we have

$$\begin{aligned} \biggl\Vert \varPhi\biggl(\frac{A+B}{2}\biggr) \bigl(\varPhi(A)\sharp\varPhi(B) \bigr)^{-1} \biggr\Vert \leq\frac{M+m}{2\sqrt{Mm}K^{\frac{1}{2}}(h_{3}^{\prime})}=\frac{K^{\frac {1}{2}}(h)}{K^{\frac{1}{2}}(h_{3}^{\prime})}. \end{aligned}$$

The proof of (2.25) is similar to (2.24), we omit the details.

This completes the proof. □

Remark 2.11

Clearly Theorem 2.10 is a refinement of (1.13).

By (1.3) and Theorem 2.8, we obtain the following refinement of inequality (1.14).

Corollary 2.12

Let\(0< m\leq M\)and\(\sqrt{\frac{M}{m}}\leq 2.314\), we have

  1. (1)

    If\(0< m\leq A\leq m_{1}^{\prime}< M_{1}^{\prime}\leq B\leq\frac{M+m}{2}\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{S^{2}(h)}{K(h_{1}^{\prime})}\varPhi ^{2}(A\sharp B), \end{aligned}$$

    where\(S(h)=\frac{h^{\frac{1}{h-1}}}{e\log h^{\frac{1}{h-1}}}\), \(K(h_{1}^{\prime})= \frac{(h_{1}^{\prime}+1)^{2}}{4h^{\prime}_{1}}\), \(h = \frac{M}{ m}\)and\(h_{1}^{\prime}= \frac{M_{1}^{\prime}}{m_{1}^{\prime}}\).

  2. (2)

    If\(0<\frac{M+m}{2}\leq A\leq m_{2}^{\prime}<M_{2}^{\prime}\leq B\leq M\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{S^{2}(h)}{K(h_{2}^{\prime})}\varPhi ^{2}(A\sharp B), \end{aligned}$$

    where\(S(h)=\frac{h^{\frac{1}{h-1}}}{e\log h^{\frac{1}{h-1}}}\), \(K(h_{2}^{\prime})= \frac{(h_{2}^{\prime}+1)^{2}}{4h_{2}^{\prime}}\), \(h = \frac{M}{m}\)and\(h_{2}^{\prime}= \frac{M_{2}^{\prime}}{m_{2}^{\prime}}\).

  3. (3)

    If\(0< m\leq A\leq m_{3}^{\prime}<\frac{M+m}{2}\leq B\leq M\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{S^{2}(h)}{K(h_{3}^{\prime})}\varPhi ^{2}(A\sharp B), \end{aligned}$$

    where\(S(h)=\frac{h^{\frac{1}{h-1}}}{e\log h^{\frac{1}{h-1}}}\), \(K(h_{3}^{\prime})= \frac{(h_{3}^{\prime}+1)^{2}}{4h_{3}^{\prime}}\), \(h = \frac{M}{m}\)and\(h_{3}^{\prime}= \frac{M+m}{2m_{3}^{\prime}}\).

  4. (4)

    If\(0< m\leq A\leq\frac{M+m}{2}<M_{3}^{\prime}\leq B\leq M\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{S^{2}(h)}{K(h_{4}^{\prime})}\varPhi ^{2}(A\sharp B), \end{aligned}$$

    where\(S(h)=\frac{h^{\frac{1}{h-1}}}{e \log h^{\frac{1}{h-1}}}\), \(K(h_{4}^{\prime})= \frac{(h_{4}^{\prime}+1)^{2}}{4h^{\prime}_{4}}\), \(h = \frac{M}{m}\)and\(h_{4}^{\prime}= \frac{2M_{3}^{\prime}}{M+m}\).

By (1.3) and Theorem 2.10, we obtain the following refinement of the inequality (1.15).

Corollary 2.13

Let\(0< m\leq M\)and\(\sqrt{\frac{M}{m}}\leq2.314\), we have

  1. (1)

    If\(0< m\leq A\leq m_{1}^{\prime}< M_{1}^{\prime}\leq B\leq\frac{M+m}{2}\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{S^{2}(h)}{K(h_{1}^{\prime})}\bigl( \varPhi (A)\sharp\varPhi( B)\bigr)^{2}, \end{aligned}$$

    where\(S(h)=\frac{h^{\frac{1}{h-1}}}{e \log h^{\frac{1}{h-1}}}\), \(K(h_{1}^{\prime})= \frac{(h_{1}^{\prime}+1)^{2}}{4h^{\prime}_{1}}\), \(h = \frac{M}{ m}\)and\(h_{1}^{\prime}= \frac{M_{1}^{\prime}}{m_{1}^{\prime}}\).

  2. (2)

    If\(0<\frac{M+m}{2}\leq A\leq m_{2}^{\prime}<M_{2}^{\prime}\leq B\leq M\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{S^{2}(h)}{K(h_{2}^{\prime})}\bigl( \varPhi (A)\sharp\varPhi( B)\bigr)^{2}, \end{aligned}$$

    where\(S(h)=\frac{h^{\frac{1}{h-1}}}{e \log h^{\frac{1}{h-1}}}\), \(K(h_{2}^{\prime})= \frac{(h_{2}^{\prime}+1)^{2}}{4h_{2}^{\prime}}\), \(h = \frac{M}{m}\)and\(h_{2}^{\prime}= \frac{M_{2}^{\prime}}{m_{2}^{\prime}}\).

  3. (3)

    If\(0< m\leq A\leq m_{3}^{\prime}<\frac{M+m}{2}\leq B\leq M\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{S^{2}(h)}{K(h_{3}^{\prime})}\bigl( \varPhi (A)\sharp\varPhi(B)\bigr)^{2}, \end{aligned}$$

    where\(S(h)=\frac{h^{\frac{1}{h-1}}}{e \log h^{\frac{1}{h-1}}}\), \(K(h_{3}^{\prime})= \frac{(h_{3}^{\prime}+1)^{2}}{4h_{3}^{\prime}}\), \(h = \frac{M}{m}\)and\(h_{3}^{\prime}= \frac{M+m}{2m_{3}^{\prime}}\).

  4. (4)

    If\(0< m\leq A\leq\frac{M+m}{2}<M_{3}^{\prime}\leq B\leq M\), then

    $$\begin{aligned} \varPhi^{2} \biggl(\frac{A+B}{2} \biggr)\leq\frac{S^{2}(h)}{K(h_{4}^{\prime})}(\varPhi (A)\sharp\varPhi( B)^{2}, \end{aligned}$$

    where\(S(h)=\frac{h^{\frac{1}{h-1}}}{e \log h^{\frac{1}{h-1}}}\), \(K(h_{4}^{\prime})= \frac{(h_{4}^{\prime}+1)^{2}}{4h^{\prime}_{4}}\), \(h = \frac{M}{m}\)and\(h_{4}^{\prime}= \frac{2M_{3}^{\prime}}{M+m}\).

References

  1. 1.

    Xue, J.: Some refinements of operator reverse AM-GM inequality. J. Inequal. Appl. 2017, Article ID 283 (2017)

    Article  Google Scholar 

  2. 2.

    Tominaga, M.: Specht’s ratio in the Young inequality. Sci. Math. Jpn. 55, 583–588 (2002)

    MathSciNet  MATH  Google Scholar 

  3. 3.

    Lin, M.: Squaring a reverse AM-GM inequality. Stud. Math. 215, 187–194 (2013)

    MathSciNet  Article  Google Scholar 

  4. 4.

    Ando, T.: Concavity of certain maps on positive definite matrices and application to Hadamard products. Linear Algebra Appl. 26, 203–241 (1979)

    MathSciNet  Article  Google Scholar 

  5. 5.

    Lin, M.: On an operator Kantorovich inequality for positive linear maps. J. Math. Anal. Appl. 402, 127–132 (2013)

    MathSciNet  Article  Google Scholar 

  6. 6.

    Fu, X., He, C.: Some operator inequalities for positive linear maps. Linear Multilinear Algebra 63, 571–577 (2015)

    MathSciNet  Article  Google Scholar 

  7. 7.

    Zhang, P.: More operator inequalities for positive linear maps. Banach J. Math. Anal. 9, 166–172 (2015)

    MathSciNet  Article  Google Scholar 

  8. 8.

    Fu, X.: An operator α-geometric mean inequality. J. Math. Inequal. 9, 947–950 (2015)

    MathSciNet  Article  Google Scholar 

  9. 9.

    Moslehian, M.S., Fu, X.: Squaring operator Polya–Szego and Diaz–Metcalf type inequalities. Linear Algebra Appl. 491, 73–82 (2016)

    MathSciNet  Article  Google Scholar 

  10. 10.

    Yang, J., Fu, X.: Squaring operator α-geometric mean inequality. J. Math. Inequal. 10, 571–575 (2016)

    MathSciNet  Article  Google Scholar 

  11. 11.

    Ali, I., Shakoor, A., Rehman, A.: More refinements of the operator reverse AM-GM inequality for positive linear maps. J. Math. Inequal. 13, 287–300 (2019)

    MathSciNet  Article  Google Scholar 

  12. 12.

    Bhatia, R., Kittaneh, F.: Notes on matrix arithmetic–geometric mean inequalities. Linear Algebra Appl. 308, 203–211 (2000)

    MathSciNet  Article  Google Scholar 

  13. 13.

    Drury, S.W.: On a question of Bhatia and Kittaneh. Linear Algebra Appl. 437, 1955–1960 (2012)

    MathSciNet  Article  Google Scholar 

  14. 14.

    Bhatia, R.: Positive Definite Matrices. Princeton University Press, Princeton (2007)

    Google Scholar 

  15. 15.

    Moradi, H.R., Omidvar, M.E., Gumus, I.H., Naseri, R.: A note on some inequalities for positive linear maps. Linear Multilinear Algebra 66, 1449–1460 (2018)

    MathSciNet  Article  Google Scholar 

Download references

Acknowledgements

The authors would like to thank the editor and the referees for their detailed and valuable suggestions for revising the manuscript.

Availability of data and materials

Not applicable.

Funding

There is no funding for this research article.

Author information

Affiliations

Authors

Contributions

All authors read and approved the final manuscript.

Corresponding author

Correspondence to Shazia Karim.

Ethics declarations

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Karim, S., Ali, I. & Mushtaq, M. Improvements of operator reverse AM-GM inequality involving positive linear maps. J Inequal Appl 2020, 73 (2020). https://doi.org/10.1186/s13660-020-02337-5

Download citation

MSC

  • 47A63
  • 47A30

Keywords

  • Operator inequalities
  • Operator reverse arithmetic-geometric mean inequality
  • Unital positive linear maps
  • Operator norm
  • Kantorovich constant