Analyzing large-scale functional data poses significant computational challenges due to high costs and substantial data storage needs. Additionally, traditional batch learning algorithms are not well-equipped to manage streaming data effectively. To address these issues, we propose a fully online learning algorithm designed for functional linear regression, which models the linear relationship between a scalar response and a functional predictor. Our approach employs Tikhonov regularization schemes within the framework of reproducing kernel Hilbert spaces (RKHS). A key feature of this fully online algorithm is its polynomially decaying regularization parameter, which adapts dynamically at each learning step, distinguishing it from the partially online algorithm that uses a fixed parameter. Within the functional linear model framework, we establish sufficient conditions for the convergence of the fully online algorithm in the RKHS norm. Additionally, we employ a capacity-independent approach to derive error bounds and almost sure convergence rates for both prediction and estimation, achieved through careful selection of step sizes and regularization parameters.

The neighbor-based method has become a powerful tool for addressing the outlier detection problem, which aims to assess the abnormality of a sample based on its compactness relative to neighboring samples. However, most existing methods primarily focus on designing various processes to identify outliers, while the contributions of different types of neighbors to the detection process have not been adequately explored. To address this gap, this article investigates the role of neighbors in existing outlier detection algorithms and introduces a taxonomy that utilizes three key components: information, neighbor, and methodology, to define hybrid methods. This taxonomy provides a framework that can inspire the development of novel neighbor-based outlier detection algorithms by combining different components from each level. Extensive comparative experiments on both synthetic and real-world datasets, including performance evaluations and case studies, demonstrate that reverse K-nearest neighbor-based methods perform well and that dynamic selection methods are particularly effective in high-dimensional spaces. Furthermore, the results confirm that strategically selecting components from this taxonomy can lead to the development of algorithms that outperform existing methods.

Outlier detection has attracted extensive attention in medical, financial, telecommunications and other fields. Although many related technologies have been proposed, most of them are faced with the problems of the neighborhood size of an object is difficult to determine and the distance in high-dimensional space is unreliable. To overcome these weaknesses, we propose a novel density-based outlier detection method that introduces the concept of Minimum the Sum of Edge Set and other related definitions in key attributes space. Based on the stability of Reverse Minimum the Sum of Edge Set, the proposed method can adaptively select the parameter representing the neighborhood size. In addition, some properties of the proposed local outlier factor are derived. Experiments on synthetic and real-world datasets demonstrate that our method is more effective than the existing outlier detection approaches.

Functional linear model is a fruitfully applied general framework for regression problems, including those with intrinsically infinite-dimensional data. Online gradient descent methods, despite their evidenced power of processing online or large-sized data, are not well studied for learning with functional data. In this paper, we study reproducing kernel-based online learning algorithms for functional data, and derive convergence rates for the expected excess prediction risk under both online and finite-horizon settings of step-sizes respectively. It is well understood that nontrivial uniform convergence rates for the estimation task depend on the regularity of the slope function. Surprisingly, the convergence rates we derive for the prediction task can assume no regularity from slope. Our analysis reveals the intrinsic difference between the estimation task and the prediction task in functional data learning.

Blockchain is making a big impact in various applications, but it is also attracting a variety of cybercrimes. In blockchain, phishing transfers the victim’s virtual currency to make huge profits through fraud, which poses a threat to the blockchain ecosystem. To avoid greater losses, Ethereum, one of the blockchain platforms, can provide information to detect phishing fraud. In this study, to effectively detect phishing nodes, we propose a phishing node detection approach as message passing based graph convolution network. We first form a transaction network through the transaction records of Ethereum and then extract the information of nodes effectively via message passing. Finally, we use a graph convolution network to classify the normal and phishing nodes. Experiments show that our method is effective and superior to other existing methods.

情感智能是机器智能的重要组成与发展方向.人工情感模型目的就是构建情感模型让机器形成情感理解与表达的系统化能力.然而现有方法在人工情感建模能力上仍显不足.针对人工情感模型构建中个性化这一关键核心因素,提出一种离散情感状态和维度空间状态互映射的方法,并基于大五人格模型和情感状态转彩模型构建一个机器个性化人工情感模型,相关实验结果对本模型的合理性提供了佐证.该研究疏通了情感空间的定义差异给机器情感建模带来的障碍,为人机个性化交互提供了一条新的解决思路,也为机器情感的理解和表达提供了可借鉴的解决方案.

In linear regression, outliers have a serious effect on the estimation of regression model parameters and the prediction of final results, so outlier detection is one of the key steps in data analysis. In this paper, we use a mean shift model and then we apply the penalty function to penalize the mean shift parameters, which is conducive to get a sparse parameter vector. We choose Sorted L1 regularization (SLOPE), which provides a convex loss function, and shows good statistical properties in parameter selection. We apply an iterative process which using gradient descent method and parameter selection at each step. Our algorithm has higher computational efficiency since the calculation of inverse matrix is avoided. Finally, we use Cross-Validation rules (CV) and Bayesian Information Criterion (BIC) criteria to fine tune the parameters, which helps our program identify outliers and obtain more robust regression coefficients. Compared with other methods, the experimental results show that our program has a fantastic performance in all aspects of outlier detection.

Iterative hard thresholding (IHT) is a beneficial tool for the recovery of sparse vectors in compressed sensing.In this study, we propose a high-accuracy distributed iterative hard thresholding algorithm (HDIHT) with explicit consideration given to the case in which noise is generated.The results of our theoretical analysis show that it is possible to cancel the noise in the HDIHT compared to the IHT.The performance of the HDIHT in the case including noise was equivalent to the classic IHT in the noise-free case.A numerical experiment is also included, and the results are in accordance with the theoretical analysis.

In this paper we study the kernel-based online gradient descent with least squares loss without an explicit regularization term. Our approach is novel by controlling the expectation of the K-norm of f(t) using an iterative process. Then we use distributed learning to improve our result.

Motivated by the recent growing interest in pairwise learning problems, we study the generalization performance of Online Pairwise lEaRning Algorithm (OPERA) in a reproducing kernel Hilbert space (RKHS) without an explicit regularization. The convergence rates established in this paper can be arbitrarily closed to O(T−12) within T iterations and largely improve the existing convergence rates for OPERA. Our novel analysis is conducted by showing an almost boundedness of the iterates encountered in the learning process with high probability after establishing an induction lemma on refining the RKHS norm estimate of the iterates.