-
How Can Large Language Models Enable Better Socially Assistive Human-Robot Interaction: A Brief Survey
Authors:
Zhonghao Shi,
Ellen Landrum,
Amy O' Connell,
Mina Kian,
Leticia Pinto-Alva,
Kaleen Shrestha,
Xiaoyuan Zhu,
Maja J Matarić
Abstract:
Socially assistive robots (SARs) have shown great success in providing personalized cognitive-affective support for user populations with special needs such as older adults, children with autism spectrum disorder (ASD), and individuals with mental health challenges. The large body of work on SAR demonstrates its potential to provide at-home support that complements clinic-based interventions deliv…
▽ More
Socially assistive robots (SARs) have shown great success in providing personalized cognitive-affective support for user populations with special needs such as older adults, children with autism spectrum disorder (ASD), and individuals with mental health challenges. The large body of work on SAR demonstrates its potential to provide at-home support that complements clinic-based interventions delivered by mental health professionals, making these interventions more effective and accessible. However, there are still several major technical challenges that hinder SAR-mediated interactions and interventions from reaching human-level social intelligence and efficacy. With the recent advances in large language models (LLMs), there is an increased potential for novel applications within the field of SAR that can significantly expand the current capabilities of SARs. However, incorporating LLMs introduces new risks and ethical concerns that have not yet been encountered, and must be carefully be addressed to safely deploy these more advanced systems. In this work, we aim to conduct a brief survey on the use of LLMs in SAR technologies, and discuss the potentials and risks of applying LLMs to the following three major technical challenges of SAR: 1) natural language dialog; 2) multimodal understanding; 3) LLMs as robot policies.
△ Less
Submitted 5 April, 2024; v1 submitted 1 April, 2024;
originally announced April 2024.
-
Designing Robot Identity: The Role of Voice, Clothing, and Task on Robot Gender Perception
Authors:
Nathaniel S. Dennler,
Mina Kian,
Stefanos Nikolaidis,
Maja Matarić
Abstract:
Perceptions of gender are a significant aspect of human-human interaction, and gender has wide-reaching social implications for robots deployed in contexts where they are expected to interact with humans. This work explored two flexible modalities for communicating gender in robots--voice and appearance--and we studied their individual and combined influences on a robot's perceived gender. We eval…
▽ More
Perceptions of gender are a significant aspect of human-human interaction, and gender has wide-reaching social implications for robots deployed in contexts where they are expected to interact with humans. This work explored two flexible modalities for communicating gender in robots--voice and appearance--and we studied their individual and combined influences on a robot's perceived gender. We evaluated the perception of a robot's gender through three video-based studies. First, we conducted a study (n=65) on the gender perception of robot voices by varying speaker identity and pitch. Second, we conducted a study (n=93) on the gender perception of robot clothing designed for two different tasks. Finally, building on the results of the first two studies, we completed a large integrative video-based study (n=273) involving two human-robot interaction tasks. We found that voice and clothing can be used to reliably establish a robot's perceived gender, and that combining these two modalities can have different effects on the robot's perceived gender. Taken together, these results inform the design of robot voices and clothing as individual and interacting components in the perceptions of robot gender.
△ Less
Submitted 30 March, 2024;
originally announced April 2024.
-
Some Generalizations of Mercer inequality and its operator extensions
Authors:
Mohsen Kian,
Zainab Peymani Mazraj
Abstract:
We study the Mercer inequality and its operator extension for superquadratic functions. In particular, we give a more general form of the Mercer inequality by replacing some constants by positive operators. As some consequences, our results produce a Jensen operator inequality for superquadratic functions. Moreover, we present some Mercer inequalities of Hermite-Hadamard's type.
We study the Mercer inequality and its operator extension for superquadratic functions. In particular, we give a more general form of the Mercer inequality by replacing some constants by positive operators. As some consequences, our results produce a Jensen operator inequality for superquadratic functions. Moreover, we present some Mercer inequalities of Hermite-Hadamard's type.
△ Less
Submitted 26 March, 2024;
originally announced March 2024.
-
Can an LLM-Powered Socially Assistive Robot Effectively and Safely Deliver Cognitive Behavioral Therapy? A Study With University Students
Authors:
Mina J. Kian,
Mingyu Zong,
Katrin Fischer,
Abhyuday Singh,
Anna-Maria Velentza,
Pau Sang,
Shriya Upadhyay,
Anika Gupta,
Misha A. Faruki,
Wallace Browning,
Sebastien M. R. Arnold,
Bhaskar Krishnamachari,
Maja J. Mataric
Abstract:
Cognitive behavioral therapy (CBT) is a widely used therapeutic method for guiding individuals toward restructuring their thinking patterns as a means of addressing anxiety, depression, and other challenges. We developed a large language model (LLM)-powered prompt-engineered socially assistive robot (SAR) that guides participants through interactive CBT at-home exercises. We evaluated the performa…
▽ More
Cognitive behavioral therapy (CBT) is a widely used therapeutic method for guiding individuals toward restructuring their thinking patterns as a means of addressing anxiety, depression, and other challenges. We developed a large language model (LLM)-powered prompt-engineered socially assistive robot (SAR) that guides participants through interactive CBT at-home exercises. We evaluated the performance of the SAR through a 15-day study with 38 university students randomly assigned to interact daily with the robot or a chatbot (using the same LLM), or complete traditional CBT worksheets throughout the duration of the study. We measured weekly therapeutic outcomes, changes in pre-/post-session anxiety measures, and adherence to completing CBT exercises. We found that self-reported measures of general psychological distress significantly decreased over the study period in the robot and worksheet conditions but not the chatbot condition. Furthermore, the SAR enabled significant single-session improvements for more sessions than the other two conditions combined. Our findings suggest that SAR-guided LLM-powered CBT may be as effective as traditional worksheet methods in supporting therapeutic progress from the beginning to the end of the study and superior in decreasing user anxiety immediately after completing the CBT exercise.
△ Less
Submitted 27 February, 2024;
originally announced February 2024.
-
Improved eigenvalue inequalities via two major subclasses of superquadratic functions
Authors:
Mohsen Kian
Abstract:
There exist two major subclasses in the class of superquadratic functions, one comprises concave and decreasing functions, while the other consists of convex and monotone increasing functions. Leveraging this distinction, we introduce eigenvalue inequalities for each case. The characteristics of these functions allow us to advance our findings in two ways: firstly, by refining existing results rel…
▽ More
There exist two major subclasses in the class of superquadratic functions, one comprises concave and decreasing functions, while the other consists of convex and monotone increasing functions. Leveraging this distinction, we introduce eigenvalue inequalities for each case. The characteristics of these functions allow us to advance our findings in two ways: firstly, by refining existing results related to eigenvalues for convex functions, and secondly, by deriving complementary inequalities for other function types. To bolster our claims, we will provide illustrative examples.
△ Less
Submitted 16 March, 2024; v1 submitted 10 July, 2023;
originally announced July 2023.
-
Matrix Inequalities between $f(A)σf(B)$ and $AσB$
Authors:
Manisha Devi,
Jaspal Singh Aujla,
Mohsen Kian,
Mohammad Sal Moslehian
Abstract:
Let $A$ and $ B$ be $n\times n$ positive definite complex matrices, let $σ$ be a matrix mean, and let $f : [0,\infty)\to [0,\infty)$ be a differentiable convex function with $f(0)=0$. We prove that $$f^{\prime}(0)(A σB)\leq \frac{f(m)}{m}(AσB)\leq f(A)σf(B)\leq \frac{f(M)}{M}(AσB)\leq f^{\prime}(M)(AσB),$$ where $m$ represents the smallest eigenvalues of $A$ and $B$ and $M$ represents the largest…
▽ More
Let $A$ and $ B$ be $n\times n$ positive definite complex matrices, let $σ$ be a matrix mean, and let $f : [0,\infty)\to [0,\infty)$ be a differentiable convex function with $f(0)=0$. We prove that $$f^{\prime}(0)(A σB)\leq \frac{f(m)}{m}(AσB)\leq f(A)σf(B)\leq \frac{f(M)}{M}(AσB)\leq f^{\prime}(M)(AσB),$$ where $m$ represents the smallest eigenvalues of $A$ and $B$ and $M$ represents the largest eigenvalues of $A$ and $B$. If $f$ is differentiable and concave, then the reverse inequalities hold. We use our result to improve some known subadditivity inequalities involving unitarily invariant norms under certain mild conditions. In particular, if $f(x)/x$ is increasing, then $$|||f(A)+f(B)|||\leq\frac{f(M)}{M} |||A+B|||\leq |||f(A+B)|||$$ holds for all $A$ and $B$ with $M\leq A+B$. Furthermore, we apply our results to explore some related inequalities. As an application, we present a generalization of Minkowski's determinant inequality.equality.
△ Less
Submitted 18 April, 2024; v1 submitted 16 February, 2023;
originally announced February 2023.
-
Generalizations of Powers--Størmer's inequality
Authors:
Mohsen Kian,
Mohammad Sal Moslehian,
Hiroyuki Osaka
Abstract:
Relating to finding possible upper bounds for the probability of error for discriminating between two quantum states, it is well-known that \begin{align*}
\mathrm{tr}(A+B) - \mathrm{tr}|A-B|\leq 2\, \mathrm{tr}\big(f(A)g(B)\big) \end{align*} holds for every positive-valued matrix monotone function $f$, where $g(x)=x/f(x)$, and all positive definite matrices $A$ and $B$. This study demonstrates t…
▽ More
Relating to finding possible upper bounds for the probability of error for discriminating between two quantum states, it is well-known that \begin{align*}
\mathrm{tr}(A+B) - \mathrm{tr}|A-B|\leq 2\, \mathrm{tr}\big(f(A)g(B)\big) \end{align*} holds for every positive-valued matrix monotone function $f$, where $g(x)=x/f(x)$, and all positive definite matrices $A$ and $B$. This study demonstrates that the set of functions satisfying this inequality includes additional elements and provides illustrative examples to support this claim. Furthermore, we present a characterization of matrix decreasing functions based on a matrix version of the above inequality.
△ Less
Submitted 16 March, 2024; v1 submitted 15 February, 2023;
originally announced February 2023.
-
Decomposition of tracial positive maps and applications in quantum information
Authors:
Ali Dadkha,
Mohsen Kian,
Mohammad Sal Moslehian
Abstract:
Every positive multilinear map between $C^*$-algebras is separately weak$^*$-continuous. We show that the joint weak$^*$-continuity is equivalent to the joint weak$^*$-continuity of the multiplications of $C^*$-algebras under consideration. We study the behavior of general tracial positive maps on properly infinite von Neumann algebras and by applying the Aron--Berner extension of multilinear maps…
▽ More
Every positive multilinear map between $C^*$-algebras is separately weak$^*$-continuous. We show that the joint weak$^*$-continuity is equivalent to the joint weak$^*$-continuity of the multiplications of $C^*$-algebras under consideration. We study the behavior of general tracial positive maps on properly infinite von Neumann algebras and by applying the Aron--Berner extension of multilinear maps, we establish that under some mild conditions every tracial positive multilinear map between general $C^*$-algebras enjoys a decomposition $Φ=\varphi_2 \circ \varphi_1$, in which $\varphi_1$ is a tracial positive linear map with the commutative range and $\varphi_2$ is a tracial completely positive map with the commutative domain. As an immediate consequence, tracial positive multilinear maps are completely positive. Furthermore, we prove that if the domain of a general tracial completely positive map $Φ$ between $C^*$-algebra is a von Neumann algebra, then $Φ$ has a similar decomposition. As an application, we investigate the generalized variance and covariance in quantum mechanics via arbitrary positive maps. Among others, an uncertainty relation inequality for commuting observables in a composite physical system is presented.
△ Less
Submitted 8 May, 2024; v1 submitted 25 February, 2022;
originally announced February 2022.
-
How type of Convexity of the Core function affects the Csiszár $f$-divergence functional
Authors:
Mohsen Kian
Abstract:
We investigate how the type of Convexity of the Core function affects the Csiszár $f$-divergence functional. A general treatment for the type of convexity has been considered and the associated perspective functions have been studied. In particular, it has been shown that when the core function is \rm{MN}-convex, then the associated perspective function is jointly \rm{MN}-convex if the two scalar…
▽ More
We investigate how the type of Convexity of the Core function affects the Csiszár $f$-divergence functional. A general treatment for the type of convexity has been considered and the associated perspective functions have been studied. In particular, it has been shown that when the core function is \rm{MN}-convex, then the associated perspective function is jointly \rm{MN}-convex if the two scalar mean \rm{M} and \rm{N} are the same. In the case where $\mathrm{M}\neq\mathrm{N}$, we study the type of convexity of the perspective function. As an application, we prove that the \textit{Hellinger distance} is jointly \rm{GG}-convex. As further applications, the matrix Jensen inequality has been developed for the perspective functions under different kinds of convexity.
△ Less
Submitted 8 January, 2021;
originally announced January 2021.
-
Jointly convex mappings related to the Lieb's functional and Minkowski type operator inequalities
Authors:
Mohsen Kian,
Yuki Seo
Abstract:
Employing the notion of operator log-convexity, we study joint concavity$/$ convexity of multivariable operator functions: $(A,B)\mapsto F(A,B)=h\left[ Φ(f(A))\ σ Ψ(g(B))\right]$, where $Φ$ and $Ψ$ are positive linear maps and $σ$ is an operator mean. As applications, we prove jointly concavity$/$convexity of matrix trace functions $\Tr\left\{ F(A,B)\right\}$. Moreover, considering positive multi-…
▽ More
Employing the notion of operator log-convexity, we study joint concavity$/$ convexity of multivariable operator functions: $(A,B)\mapsto F(A,B)=h\left[ Φ(f(A))\ σ Ψ(g(B))\right]$, where $Φ$ and $Ψ$ are positive linear maps and $σ$ is an operator mean. As applications, we prove jointly concavity$/$convexity of matrix trace functions $\Tr\left\{ F(A,B)\right\}$. Moreover, considering positive multi-linear mappings in $F(A,B)$, our study of the joint concavity$/$ convexity of $(A_1,\cdots,A_k)\mapsto h\left[ Φ(f(A_1),\cdots,f(A_k))\right]$ provides some generalizations and complement to results of Ando and Lieb concerning the concavity$/$ convexity of maps involving tensor product. In addition, we present Minkowski type operator inequalities for a unial positive linear map, which is an operator version of Minkowski type matrix trace inequalities under a more general setting than Carlen and Lieb, Bekjan, and Ando and Hiai.
△ Less
Submitted 24 October, 2020;
originally announced October 2020.
-
Klein's trace inequality and superquadratic trace functions
Authors:
Mohsen Kian,
Mohammad W. Alomari
Abstract:
We show that if $f$ is a non-negative superquadratic function, then $A\mapsto\mathrm{Tr}f(A)$ is a superquadratic function on the matrix algebra. In particular,
\begin{align*}
\tr f\left( {\frac{A + B}{2}} \right) +\tr f\left(\left| {\frac{A - B}{2}}\right|\right) \leq \frac{{\tr {f\left( A \right)} + \tr {f\left( B \right)} }}{2} \end{align*}
holds for all positive matrices $A,B$.
In addi…
▽ More
We show that if $f$ is a non-negative superquadratic function, then $A\mapsto\mathrm{Tr}f(A)$ is a superquadratic function on the matrix algebra. In particular,
\begin{align*}
\tr f\left( {\frac{A + B}{2}} \right) +\tr f\left(\left| {\frac{A - B}{2}}\right|\right) \leq \frac{{\tr {f\left( A \right)} + \tr {f\left( B \right)} }}{2} \end{align*}
holds for all positive matrices $A,B$.
In addition, we present a Klein's inequality for superquadratic functions as $$
\mathrm{Tr}[f(A)-f(B)-(A-B)f'(B)]\geq \mathrm{Tr}[f(|A-B|)] $$ for all positive matrices $A,B$. It gives in particular an improvement of the Klein's inequality for non-negative convex function.
As a consequence, some variants of the Jensen trace inequality for superquadratic functions have been presented.
△ Less
Submitted 27 January, 2020;
originally announced January 2020.
-
Asymmetric Choi--Davis inequalities
Authors:
Mohsen Kian,
M. S. Moslehian,
R. Nakamoto
Abstract:
Let $Φ$ be a unital positive linear map and let $A$ be a positive invertible operator. We prove that there exist partial isometries $U$ and $V$ such that \[ |Φ(f(A))Φ(A)Φ(g(A))|\leq U^*Φ(f(A)Ag(A))U \] and \[\left|Φ\left(f(A)\right)^{-r}Φ(A)^rΦ\left(g(A)\right)^{-r}\right|\leq V^*Φ\left(f(A)^{-r}A^rg(A)^{-r}\right)V\] hold under some mild operator convex conditions and some positive numbers $r$. F…
▽ More
Let $Φ$ be a unital positive linear map and let $A$ be a positive invertible operator. We prove that there exist partial isometries $U$ and $V$ such that \[ |Φ(f(A))Φ(A)Φ(g(A))|\leq U^*Φ(f(A)Ag(A))U \] and \[\left|Φ\left(f(A)\right)^{-r}Φ(A)^rΦ\left(g(A)\right)^{-r}\right|\leq V^*Φ\left(f(A)^{-r}A^rg(A)^{-r}\right)V\] hold under some mild operator convex conditions and some positive numbers $r$. Further, we show that if $f^2$ is operator concave, then $$ |Φ(f(A))Φ(A)|\leq Φ(Af(A)).$$ In addition, we give some counterparts to the asymmetric Choi--Davis inequality and asymmetric Kadison inequality. Our results extend some inequalities due to Bourin--Ricard and Furuta.
△ Less
Submitted 27 January, 2020;
originally announced January 2020.
-
Matricial Radius: A Relation of Numerical radius with Matricial Range
Authors:
Mohsen Kian,
Mahdi Dehghani,
Mostafa Sattari
Abstract:
It has been shown that if $T$ is a complex matrix, then {\small\begin{align*} ω(T)&=\frac{1}{n}\sup\left\{|\mathrm{Tr}\ X|;\ X\in W^n(T)\right\}\\ &=\frac{1}{n}\sup\left\{\|X\|_1;\ X\in W^n(T)\right\}\\ &= \sup\left\{ ω(X);\ X\in W^n(T)\right\} \end{align*} }
where $n$ is a positive integer, $ω(T)$ is the numerical radius and $W^n(T)$ is the $n$'th matricial range of $T$.
It has been shown that if $T$ is a complex matrix, then {\small\begin{align*} ω(T)&=\frac{1}{n}\sup\left\{|\mathrm{Tr}\ X|;\ X\in W^n(T)\right\}\\ &=\frac{1}{n}\sup\left\{\|X\|_1;\ X\in W^n(T)\right\}\\ &= \sup\left\{ ω(X);\ X\in W^n(T)\right\} \end{align*} }
where $n$ is a positive integer, $ω(T)$ is the numerical radius and $W^n(T)$ is the $n$'th matricial range of $T$.
△ Less
Submitted 25 November, 2019;
originally announced November 2019.
-
Variants of Ando--Hiai type inequalities for deformed means and applications
Authors:
Mohsen Kian,
M. S. Moslehian,
Yuki Seo
Abstract:
For an $n$-tuple of positive invertible operators on a Hilbert space, we present some variants of Ando--Hiai type inequalities for deformed means from an $n$-variable operator mean by an operator mean, which is related to the information monotonicity of a certain unital positive linear map. As an application, we investigate the monotonicity of the power mean from the deformed mean in terms of the…
▽ More
For an $n$-tuple of positive invertible operators on a Hilbert space, we present some variants of Ando--Hiai type inequalities for deformed means from an $n$-variable operator mean by an operator mean, which is related to the information monotonicity of a certain unital positive linear map. As an application, we investigate the monotonicity of the power mean from the deformed mean in terms of the generalized Kantorovich constants under the operator order. Moreover, we improve the norm inequality for the operator power means related to the Log-Euclidean mean in terms of the Specht ratio.
△ Less
Submitted 25 November, 2019;
originally announced November 2019.
-
Positivity of $2\times 2$ block matrices of operators
Authors:
Mohammad Sal Moslehian,
Mohsen Kian,
Qingxiang Xu
Abstract:
We review some of the significant generalizations and applications of the celebrated Douglas theorem on the equivalence of factorization, range inclusion, and majorization of operators. We then apply it to find a characterization of the positivity of $2\times 2$ block matrices of operators in Hilbert spaces and finally describe the nature of such block matrices and provide several ways for showing…
▽ More
We review some of the significant generalizations and applications of the celebrated Douglas theorem on the equivalence of factorization, range inclusion, and majorization of operators. We then apply it to find a characterization of the positivity of $2\times 2$ block matrices of operators in Hilbert spaces and finally describe the nature of such block matrices and provide several ways for showing their positivity.
△ Less
Submitted 18 April, 2019;
originally announced April 2019.
-
Power means of probability measures and Ando-Hiai inequality
Authors:
Mohsen Kian,
Mohammad Sal Moslehian
Abstract:
Let $μ$ be a probability measure of compact support on the set $\mathbb{P}_n$ of all positive definite matrices, let $t\in(0,1]$, and let $P_t(μ)$ be the unique positive solution of $X=\int_{\mathbb{P}_n}X\sharp_t Z dμ(Z)$. In this paper, we show that
$$ P_t(μ)\leq I\quad \Longrightarrow\quad P_{\frac{t}{p}}(ν)\leq P_t(μ)$$ for every $p\geq1$, where $ν(Z)=μ(Z^{1/p})$. This provides an extension…
▽ More
Let $μ$ be a probability measure of compact support on the set $\mathbb{P}_n$ of all positive definite matrices, let $t\in(0,1]$, and let $P_t(μ)$ be the unique positive solution of $X=\int_{\mathbb{P}_n}X\sharp_t Z dμ(Z)$. In this paper, we show that
$$ P_t(μ)\leq I\quad \Longrightarrow\quad P_{\frac{t}{p}}(ν)\leq P_t(μ)$$ for every $p\geq1$, where $ν(Z)=μ(Z^{1/p})$. This provides an extension of the Ando--Hiai inequality for matrix power means. Moreover, we prove that if $Φ:\mathbb{M}_n\to\mathbb{M}_m$ is a unital positive linear map, then $Φ(P_t(μ))\leq P_t(ν)$ for all $t\in[-1,1]\backslash\{0\}$, where $ν$ is a certain measure.
△ Less
Submitted 22 September, 2019; v1 submitted 11 June, 2018;
originally announced June 2018.
-
Interpolating operator Jensen-type inequalities for log-convex and superquadratic functions
Authors:
Mojtaba Bakherad,
Mohsen Kian,
Mario Krnic,
Seyyed Alireza Ahmadi
Abstract:
Motivated by some recently established operator Jensen-type inequalities related to a usual convexity, in the present paper we derive several more accurate operator Jensen-type inequalities for certain subclasses of convex functions. More precisely, we obtain interpolating series of Jensen-type inequalities for log-convex and non-negative superquadratic functions. In particular, we obtain the corr…
▽ More
Motivated by some recently established operator Jensen-type inequalities related to a usual convexity, in the present paper we derive several more accurate operator Jensen-type inequalities for certain subclasses of convex functions. More precisely, we obtain interpolating series of Jensen-type inequalities for log-convex and non-negative superquadratic functions. In particular, we obtain the corresponding refinements of the Jensen-Mercer operator inequality for such classes of functions.
△ Less
Submitted 10 May, 2018;
originally announced May 2018.
-
A refinement of Hardy Inequality via superquadratic function
Authors:
Mohsen Kian,
M. Rostamian Delavar
Abstract:
A refinement of the Hardy inequality has been presented by use of superquadratic function.
A refinement of the Hardy inequality has been presented by use of superquadratic function.
△ Less
Submitted 16 May, 2017;
originally announced May 2017.
-
C*-Convexity of Norm Unit Balls
Authors:
Mohsen Kian
Abstract:
We determine those norms on B(H) whose unit ball is C*-convex. We call them M-norms and show that the class of M-norms less than a given norm enjoys a maximum element. These minimum and maximum elements will be determined in some cases. Finally, we give a constructive result to obtain M-norms on B(H).
We determine those norms on B(H) whose unit ball is C*-convex. We call them M-norms and show that the class of M-norms less than a given norm enjoys a maximum element. These minimum and maximum elements will be determined in some cases. Finally, we give a constructive result to obtain M-norms on B(H).
△ Less
Submitted 11 April, 2016;
originally announced April 2016.
-
Epigraph of Operator Functions
Authors:
Mohsen Kian
Abstract:
It is known that a real function $f$ is convex if and only if the set $$\mathrm{E}(f)=\{(x,y)\in\mathbb{R}\times\mathbb{R};\ f(x)\leq y\},$$ the epigraph of $f$ is a convex set in $\mathbb{R}^2$. We state an extension of this result for operator convex functions and $C^*$-convex sets as well as operator log-convex functions and $C^*$-log-convex sets. Moreover, the $C^*$-convex hull of a Hermitian…
▽ More
It is known that a real function $f$ is convex if and only if the set $$\mathrm{E}(f)=\{(x,y)\in\mathbb{R}\times\mathbb{R};\ f(x)\leq y\},$$ the epigraph of $f$ is a convex set in $\mathbb{R}^2$. We state an extension of this result for operator convex functions and $C^*$-convex sets as well as operator log-convex functions and $C^*$-log-convex sets. Moreover, the $C^*$-convex hull of a Hermitian matrix has been represented in terms of its eigenvalues.
△ Less
Submitted 17 December, 2015;
originally announced December 2015.
-
Developed Matrix inequalities via Positive Multilinear Mappings
Authors:
Mahdi Dehghani,
Mohsen Kian,
Yuki Seo
Abstract:
Utilizing the notion of positive multilinear mappings, we give some matrix inequalities. In particular, Choi--Davis--Jensen and Kantorovich type inequalities including positive multilinear mappings are presented.
Utilizing the notion of positive multilinear mappings, we give some matrix inequalities. In particular, Choi--Davis--Jensen and Kantorovich type inequalities including positive multilinear mappings are presented.
△ Less
Submitted 8 December, 2015;
originally announced December 2015.
-
Operator log-convex functions and f-divergence functional
Authors:
Mohsen Kian
Abstract:
We present a characterization of operator log-convex functions by using positive linear mappings. Moreover, we study the non-commutative f-divergence functional of operator log-convex functions. In particular, we prove that f is operator log-convex if and only if the non-commutative f-divergence functional is operator log-convex in its first variable and operator convex in its second variable.
We present a characterization of operator log-convex functions by using positive linear mappings. Moreover, we study the non-commutative f-divergence functional of operator log-convex functions. In particular, we prove that f is operator log-convex if and only if the non-commutative f-divergence functional is operator log-convex in its first variable and operator convex in its second variable.
△ Less
Submitted 25 August, 2014; v1 submitted 31 May, 2013;
originally announced May 2013.
-
Operator inequalities of Jensen type
Authors:
M. S. Moslehian,
J. Micic,
M. Kian
Abstract:
We present some generalized Jensen type operator inequalities involving sequences of self-adjoint operators. Among other things, we prove that if $f:[0,\infty) \to \mathbb{R}$ is a continuous convex function with $f(0)\leq 0$, then {equation*} \sum_{i=1}^{n} f(C_i) \leq f(\sum_{i=1}^{n}C_i)-δ_f\sum_{i=1}^{n}\widetilde{C}_i\leq f(\sum_{i=1}^{n}C_i) {equation*} for all operators $C_i$ such that…
▽ More
We present some generalized Jensen type operator inequalities involving sequences of self-adjoint operators. Among other things, we prove that if $f:[0,\infty) \to \mathbb{R}$ is a continuous convex function with $f(0)\leq 0$, then {equation*} \sum_{i=1}^{n} f(C_i) \leq f(\sum_{i=1}^{n}C_i)-δ_f\sum_{i=1}^{n}\widetilde{C}_i\leq f(\sum_{i=1}^{n}C_i) {equation*} for all operators $C_i$ such that $0 \leq C_i\leq M \leq \sum_{i=1}^{n} C_i $ \ $(i=1,...,n)$ for some scalar $M\geq0$, where $ \widetilde{C_i} = 1/2 - |\frac{C_i}{M}- 1/2 |$ and $δ_f = f(0)+f(M) - 2 f(\frac{M}{2})$.
△ Less
Submitted 31 March, 2013;
originally announced April 2013.
-
Non-commutative f-divergence functional
Authors:
Mohammad Sal Moslehian,
Mohsen Kian
Abstract:
We introduce the non-commutative $f$-divergence functional $Θ(\widetilde{A},\widetilde{B}):=\int_TB_t^{\frac{1}{2}}f\left(B_t^{-\frac{1}{2}} A_tB_t^{-\frac{1}{2}}\right)B_t^{\frac{1}{2}}dμ(t)$ for an operator convex function $f$, where $\widetilde{A}=(A_t)_{t\in T}$ and $\widetilde{B}=(B_t)_{t\in T}$ are continuous fields of Hilbert space operators and study its properties. We establish some relat…
▽ More
We introduce the non-commutative $f$-divergence functional $Θ(\widetilde{A},\widetilde{B}):=\int_TB_t^{\frac{1}{2}}f\left(B_t^{-\frac{1}{2}} A_tB_t^{-\frac{1}{2}}\right)B_t^{\frac{1}{2}}dμ(t)$ for an operator convex function $f$, where $\widetilde{A}=(A_t)_{t\in T}$ and $\widetilde{B}=(B_t)_{t\in T}$ are continuous fields of Hilbert space operators and study its properties. We establish some relations between the perspective of an operator convex function $f$ and the non-commutative $f$-divergence functional. In particular, an operator extension of Csiszár's result regarding $f$-divergence functional is presented. As some applications, we establish a refinement of the Choi--Davis--Jensen operator inequality, obtain some unitarily invariant norm inequalities and give some results related to the Kullback--Leibler distance.
△ Less
Submitted 30 January, 2013;
originally announced January 2013.
