Search

Consider the recent upswing in interest around multi-view clustering procedures. Such methods aim to boost clustering efficiency by leveraging information from numerous perspectives. Much research has been devoted to tensorial representation to exploit high-order correlations underlying disparate views while preserving the local geometric structure inside each view. Our research introduces a novel multi-view clustering approach. This approach creates a 3rd-order tensor, assimilating features from all perspectives. We use the t-product in the tensor space to generate the self-representation tensor from the tensorial data. We incorporate the ℓ 1 -based sliced-Laplacian regularization to increase our model's resilience and introduce a fresh column-wise sparse norm: the ℓ 2 , p -norm with 0 < p < 1. This norm displays attributes of invariance, continuity, and differentiability. We present a closed-form answer to the ℓ 2 , p -regularized shrinkage problem, broadening its relevance to other generalized problems. Simultaneously, we propose a tensorial arctan -function as an improved surrogate for the tensor rank. This function has proven more proficient at assessing consistency across multiple viewpoints. By integrating these two components, we formulate an effective algorithm that refines our suggested model, ensuring that the constructed sequence gravitates toward the stationary KKT point. Our team conducts extensive experiments on various datasets to evaluate our model's effectiveness, spanning diverse situations and scales. Results from these experiments emphasize that our approach establishes a novel performance standard. • We integrate ℓ 1 -sliced Laplacian with ℓ 2 , p regularization for multi-view clustering. It reduces redundancy and enhances outlier resilience, achieving an efficient, innovative solution. • An algorithm with augmented Lagrange multipliers tackles multi-view clustering, providing solvable sub-problems with closed-form solutions. • Within the tensor framework adept at leveraging intrinsic cluster structures, our algorithm approaches an optimal Karush–Kuhn–Tucker (KKT) point. • Extensive experiments on four datasets demonstrate our technique's superiority over contemporary methods through thorough comparisons. [ABSTRACT FROM AUTHOR]

• A novel modified Jaya algorithm is developed to identify structural damage. • A multi-sample objective function embedding L 0.5 sparsity regularization is presented. • The coupling effects of three types of uncertainties are considered during the calculation. • Satisfactory identifications can be obtained by the proposed method. This study develops an improved Jaya algorithm for structural damage identification considering various uncertainties using vibration data. Most studies consider uncertainties, such as measurement noise, modeling uncertainties, and temperature variations separately but few on their coupling effects. On the other hand, the Jaya algorithm may trap local minimums in optimizing complex objective functions. Therefore, a novel modified Jaya algorithm is developed by integrating the one-step K-means clustering, Hooke–Jeeves pattern search, and linear population reduction strategies. The three modification strategies greatly improve the global optimization performance of the standard Jaya algorithm, which is verified by a series of high-dimension test functions. An L 0.5 regularization is applied to improve the ill-posedness of the damage identification problem and ensure the sparsity of the solution. Numerical and experimental studies on two structures show that the proposed algorithm can identify the structural damage accurately, even considering the coupling effects of various types of uncertainties. [ABSTRACT FROM AUTHOR]