Multi-source domain adaptation (MDA) aims to transfer knowledge from multiple source domains an unlabeled target domain. MDA is a challenging task due the severe shift, which not only exists between and but also among diverse sources. Prior studies on either estimate mixed distribution of or combine single-source models, few them delve into relevant information domains. For this reason, we propose novel approach, termed Pseudo Target for (PTMDA). Specifically, PTMDA maps each group...

Traditional financial data mining and information disclosure research faces problems such as huge volume, uneven quality, untimely disclosure. This paper aims to use fuzzy logic algorithms process data, improve efficiency, explore new methods, design an evaluation model based on algorithms. The experiment constructed a adaptive neural inference system (ANFIS), optimized the parameters using genetic algorithm (FGA), used comprehensive form content with help of logical reasoning. results show...

Suppose $A - BXB^*$, BX X^*B^*$, and + X^*B^*$ are three linear matrix expressions over the field of complex numbers, where A is an Hermitian or skew‐Hermitian matrix. In this paper, we consider how to choose X such that BXB^*$ have maximal minimal possible ranks, \pm attain ranks. Some applications solutions equations with symmetric patterns also given.

In an existing brain-computer interface (BCI) based on code modulated visual evoked potentials (c-VEP), a method with which to increase the number of targets without increasing length has not yet been established. this paper, novel c-VEP BCI paradigm, namely, grouping modulation different codes that have good autocorrelation and crosscorrelation properties, is presented information transfer rate (ITR). All stimulus are divided into several groups each group by distinct pseudorandom binary...

Abstract Through a Hermitian‐type (skew‐Hermitian‐type) singular value decomposition for pair of matrices ( A, B ) introduced by Zha Linear Algebra Appl. 1996; 240 :199–205), where A is Hermitian (skew‐Hermitian), we show how to find (skew‐Hermitian) matrix X such that the expressions − BX ± * achieve their maximal and minimal possible ranks, respectively. For consistent equations = , give general solutions through two kinds generalized decompositions. As applications linear model { y β, σ 2...

Researches on ranks of matrix expressions have posed a number challenging questions, one which is concerned with simultaneous decompositions several given matrices. In this paper, we construct decomposition to triplet (A, B, C), where A=±A*. Through the decomposition, derive canonical form for A−BXB*−CYC* and then solve two conjectures maximal minimal possible respect X=±X* Y=±Y*. As an application, sufficient necessary condition equation BXB* + CYC*=A pair Hermitian solutions, give general...

The number of selectable targets is one the main factors that affect performance a brain-computer interface (BCI). Most existing code modulated visual evoked potential (c-VEP) based BCIs use single pseudorandom binary sequence and its circularly shifting sequences to modulate different stimulus targets, making limited by length modulation codes. This paper proposes novel paradigm for c-VEP BCIs, which divides into four target groups each group are unique Based on paradigm, four-group BCI...

Necessary and sufficient conditions are established for a group of mixed-type reverseorder laws generalized inverses triple matrixproduct to hold. Some applications thereverse-order the sum two matrices also given.

Recent studies have adopted pre-trained language models, such as CodeT5 and CodeGPT, for automated program generation tasks like code generation, repair, translation. Numerous model-based approaches been proposed evaluated on various benchmark datasets, demonstrating promising performance. However, there is still uncertainty about the reliability of these particularly their realistic ability to consistently transform sequences. This raises question: are techniques sufficiently trustworthy...

Android has become the predominant smartphone operating system, with a rapidly evolving ecosystem that requires app developers to frequently update their apps maintain quality, security, and compatibility. While deep learning made significant strides in various software engineering tasks, including automated code updates, existing methods are not specifically tailored for apps, potential of pre-trained Language Models Code (CodeLMs) updating remains unexplored. In this paper, we present...

Ordinary least squares estimator (OLSE), best linear unbiased (BLUE), and predictor (BLUP) in the general model with new observations are generalized to multivariate model. The fundamental equations of BLUE BLUP derived by two methods, including vectorization method projection method. By using matrix rank method, some results BLUE-sufficiency, BLUP-sufficiency, equality OLSE, BLUE, given

Unsupervised domain adaptation (UDA) is one of the prominent tasks transfer learning, and it provides an effective approach to mitigate distribution shift between labeled source unlabeled target domain. Prior works mainly focus on aligning marginal distributions or estimated class-conditional distributions. However, joint dependency among feature label crucial for task not fully exploited. To address this problem, we propose Bures Joint Distribution Alignment (BJDA) algorithm which directly...

To observe the clinical value of 3D printing technology assisted surgery combined with early postoperative comprehensive rehabilitation in elderly patients intertrochanteric fractures.