Searcharxiv⌕ Search

arXiv subjects

Cailing Yao

Publications and source records attributed to Cailing Yao.

8 recordsLinked to original sources

Quaternion Matrix Tuples with Common Invariant Subspaces: Dimension and Nonuniqueness

Let $M_n(\mathbb{H})$ denote the algebra of $n\times n$ quaternion matrices. We study common invariant right subspaces of tuples in $M_n(\mathbb{H})^r$ and determine exact real semialgebraic dimensions for two associated sets. If $n_1,\ldots,n_s$ are positive integers, $n=n_1+\cdots+n_s$, and $q=\sum_{1\leq a<b\leq s}n_an_b$, the set of tuples preserving at least one nested chain with successive increments $n_1,\ldots,n_s$ has dimension $4rn^2-4(r-1)q$. For $r\geq2$ and $1\leq k<n$, the set of tuples with at least two distinct common $k$-dimensional right invariant subspaces has dimension $4rn^2-4(r-1)(k(n-k)+n-1)$. The proofs use finite right-module coordinates and a common local-rank argument.

math.RA↗

Unitary similarity for dual complex matrices and dual quaternion matrices

We study unitary similarity of dual complex matrices and dual quaternion matrices. First, we characterize unitary similarity in terms of the existence of a skew-Hermitian solution to a Sylvester-type matrix equation. For $2\times2$ dual complex matrices, we obtain a complete three-trace criterion when the standard part is nonnormal and analyze the limitations of trace-word criteria when the standard part is normal. We then prove, for every $n\geq1$, that two $n\times n$ dual quaternion matrices are unitarily similar if and only if their dual complex adjoint matrices are unitarily similar, and we extend this result to finite families. As a scalar consequence, we derive a finite Trace-word criterion when the standard part is nonreal and exhibit its failure when the standard part is real.

math.RA↗

Complex Representatives, Similarity Invariants, and Iterates in $\bf{U(1,1;\mathbb{H})}$

We study the similarity classes of matrices in $U(1,1;\mathbb H)$ and the iterates of their associated quaternionic Möbius transformations. Building on the known elliptic case, we show that every parabolic or loxodromic element of $U(1,1;\mathbb H)$ is similar, via a matrix in $U(1,1;\mathbb H)$, to a complex matrix. Explicit complex representatives and matrices realizing these similarities are constructed for the different types. We then introduce the invariants $\operatorname{Tr}(T), Δ(T), η(T), κ(T)$, and prove that they form a complete system of similarity invariants for matrices in $U(1,1;\mathbb H)$. Finally, using the complex representation of quaternionic matrices and Chebyshev polynomials, we derive a fourth-order recurrence for a real sequence $(q_n)$ satisfying $Δ(T^n)=q_n^2Δ(T)$. Together with recurrence relations for the relevant matrix entries, this identity yields a complete classification of the quaternionic Möbius transformation associated with $T^n$, including the degenerate cases in which $Δ(T^n)=0$.

math.FA↗

The $ϕ$-conjugation of quaternionic matrices and generalized Autonne-Takagi factorization

Let $ϕ$ be a quaternion of modulus $1$. In this article, we study some topics related to $ϕ$-conjugation for quaternionic matrices, including $ϕ$-Hermitian matrices, $ϕ$-conjugate normal matrices, unitary $ϕ$-congruence and $ϕ$-HSH decomposition (decomposition of a $ϕ$-Hermitian matrix and a skew $ϕ$-Hermitian matrix). In particular, we generalize the Autonne-Takagi factorization of quaternion $ϕ$-Hermitian matrices for all unit quaternion $ϕ$. This gives an affirmative answer to a problem proposed by R. Horn and F. Zhang in the paper ``A generalization of the complex Autonne-Takagi factorization to quaternion matrices, Linear Multilinear A. 60: 1239--1244, 2012''.

math.RA↗

Diagonalization of skew-Hermitian dual quaternion matrices and $ϕ$-Hermitian dual quaternion matrices

In this article, we present the diagonalization of skew-Hermitian dual quaternion matrices in the sense of unitary equivalence. For a unit quaternion $ϕ$, we also obtain the diagonalization result of $ϕ$-Hermitian dual quaternion matrices in the sense of unitary $ϕ$-congruence, which generalizes the Autonne--Takagi factorization of symmetric complex matrices and $ϕ$-Hermitian quaternion matrices.

math.RA↗

$\boldsymbol{i}$-conjugate for quaternionic matrices and related properties

Motivated by the result that a complex $n\times n$ matrix $A$ being unitarily equivalent to a real matrix, we extend the conclusion to the quaternion skew field in this paper, we present a necessary and sufficient condition for that a quaternion $n\times n$ matrix $A$ is unitarily equivalent to a complex matrix. To state the truth more clearly, we put forward the concept which we call $\boldsymbol{i}$-conjugate. Furthermore, we study the concepts related to $\boldsymbol{i}$-conjugate and their properties, such as unitary $\boldsymbol{i}$-congruence, $\boldsymbol{i}$-conjugate normality and $\boldsymbol{i}$-Hermicity in $M_{n}(\mathbb{H})$ as generalizations of the conventional unitary congruence, conjugate normality and Hermicity of matrices in $M_{n}(\mathbb{C})$. Finally, we present a new type of polar decomoposition of quaternion matrices.

math.RA↗

G-LoG Bi-filtration for Medical Image Classification

Building practical filtrations on objects to detect topological and geometric features is an important task in the field of Topological Data Analysis (TDA). In this paper, leveraging the ability of the Laplacian of Gaussian operator to enhance the boundaries of medical images, we define the G-LoG (Gaussian-Laplacian of Gaussian) bi-filtration to generate the features more suitable for multi-parameter persistence module. By modeling volumetric images as bounded functions, then we prove the interleaving distance on the persistence modules obtained from our bi-filtrations on the bounded functions is stable with respect to the maximum norm of the bounded functions. Finally, we conduct experiments on the MedMNIST dataset, comparing our bi-filtration against single-parameter filtration and the established deep learning baselines, including Google AutoML Vision, ResNet, AutoKeras and auto-sklearn. Experiments results demonstrate that our bi-filtration significantly outperforms single-parameter filtration. Notably, a simple Multi-Layer Perceptron (MLP) trained on the topological features generated by our bi-filtration achieves performance comparable to complex deep learning models trained on the original dataset.

cs.CV↗

Some invariants of $U(1,1;\mathbb{H})$ and diagonalization

Denote by $\mathbb{H}$ the set of all quaternions. We are interested in the group $U(1,1;\mathbb{H})$, which is a subgroup of $2\times 2$ quaternionic matrix group and is sometimes called $Sp(1,1)$. As well known, $U(1,1;\mathbb{H})$ corresponds to the quaternionic Möbius transformations on the unit ball in $\mathbb{H}$. In this article, some similar invariants on $U(1,1;\mathbb{H})$ are discussed. Our main result shows that each matrix $T\in U(1,1;\mathbb{H})$, which corresponds to an elliptic quaternionic Möbius transformation $g_T(z)$, could be $U(1,1;\mathbb{H})$-similar to a diagonal matrix.

math.RA↗