Unsupervised hashing has been extensively applied in large-scale multi-modal retrieval by mapping original data from heterogeneous modalities into unified binary codes. However, there still remain ...challenges especially how to balance the individual modality-specific representations and common representation preserving intrinsic linkages among heterogeneous modalities. In this paper, we propose a novel fast Unsupervised Multi-modal Hashing based on Piecewise Learning, denoted as UMHPL, to deal with the mentioned issue. Initially, we formulate the problem as matrix factorization to derive the individual modality-specific latent representations and common latent representation with consensus matrices in a brief time. To maintain the integrality of multi-modal data, we integrate them by adaptive weight factors and nuclear norm minimization. Subsequently, we establish a connection between the individual modality-specific latent representations and common latent representation based on the piecewise hash learning framework to reinforce the discriminative competency of model, which leads the hash codes more compact. Finally, an effective discrete optimization algorithm in mathematical logic and functional analysis is proposed. Comprehensive experiments on Wiki, MIRFlirck, NUS-WIDE, and MSCOCO datasets demonstrate the superior performance of UMHPL to state-of-the-art hashing methods.
Emperor Penguin Optimizer (EPO) is a metaheuristic algorithm which is recently developed and illustrates the emperor penguin’s huddling behaviour. However, the original version of the EPO will fix ...issues that are continuing in fact but not discrete. The eight separate EPO variants have been provided in this article. Four transfer features, s-shaped and v-shaped, that are used in order to map the search space into a separate research space are considered in the proposed algorithm. The output of the proposed algorithm is validated using 25 standard benchmark functions. It also analyses the statistical sense of the proposed algorithm. Experimental findings and comparisons suggest that the proposed algorithm performs better than other algorithms. The solution also applies to the issue of feature selection. The findings reveal the supremacy of the binary emperor penguin optimization algorithm.
Hashing methods have been extensively applied to efficient multimedia data indexing and retrieval on account of the explosion of multimedia data. Cross-modal hashing usually learns binary codes by ...mapping multi-modal data into a common Hamming space. Most supervised methods utilize relation information like class labels as pairwise similarities of cross-modal data pair to narrow intra-modal and inter-modal gap. In this paper, we propose a novel supervised cross-modal hashing method dubbed Subspace Relation Learning for Cross-modal Hashing (SRLCH), which exploits relation information of labels in semantic space to make similar data from different modalities closer in the low-dimension Hamming subspace. SRLCH preserves the modality relationships, the discrete constraints and nonlinear structures, while admitting a closed-form binary codes solution, which effectively enhances the training efficiency. An iterative alternative optimization algorithm is developed to simultaneously learn both hash functions and unified binary codes. With these binary codes and hash functions, we can index multimedia data and search them in an efficient way. Evaluations in two cross-modal retrieval tasks on several widely-used datasets show that the proposed SRLCH outperforms most cross-modal hashing methods. Theoretical analysis also illustrates reasons for our method's promotion in subspace relation learning.
The ship base is a structure that connects the equipment to the hull and may play a role in restraining and isolating the dynamic load. Adding damping on the base to improve the vibration isolation ...performance is an important measure to control ship vibration. In this research, the energy transfer route and vector cloud of the ship base were analyzed employing the power flow theory, and then the placement of the particle damper was determined. Through the discrete optimization of different particle parameters including the particle material, diameter and filling rate, the best vibration reduction effect was acquired. The simulation and experiment results show that the particle damping has obvious damping effect, and the steel particle has better damping effect than the lead particle and the aluminum particle. The change of particle filling rate influences the vibration characteristics, and the best effect is achieved when the filling rate is 82%. The vibration reduction performance relies strongly on particle diameters, and they all exert obvious vibration suppression effect at the peak acceleration admittance. The proposed discrete optimization strategy effectively saves experiment cost, and the presented particle damper may be traded as an optional scheme in vibration reduce treatment of ship base.
•The visualized power flow analysis method is used to determine the area of damping treatment.•A discrete optimization method is proposed to optimize the matching of particle damping parameters (Fe, 4mm, 82%).
The design of evolutionary algorithms has typically been focused on efficiently solving a single optimization problem at a time. Despite the implicit parallelism of population-based search, no ...attempt has yet been made to multitask, i.e., to solve multiple optimization problems simultaneously using a single population of evolving individuals. Accordingly, this paper introduces evolutionary multitasking as a new paradigm in the field of optimization and evolutionary computation. We first formalize the concept of evolutionary multitasking and then propose an algorithm to handle such problems. The methodology is inspired by biocultural models of multifactorial inheritance, which explain the transmission of complex developmental traits to offspring through the interactions of genetic and cultural factors. Furthermore, we develop a cross-domain optimization platform that allows one to solve diverse problems concurrently. The numerical experiments reveal several potential advantages of implicit genetic transfer in a multitasking environment. Most notably, we discover that the creation and transfer of refined genetic material can often lead to accelerated convergence for a variety of complex optimization functions.
Deep hashing has been widely used for large-scale cross-modal retrieval benefited from the low storage cost and fast search speed. However, most existing deep supervised methods only preserve the ...instance-pairwise relationship supervised by the semantic similarity matrix, which always inufficient heterogeneous correlation. Thus, we propose the Deep Discrete Cross-Modal Hashing with Multiple Supervision (DDCHms) to further enhance the semantic consistency of heterogeneous modalities. It improves the performance of semantic information retrieval with the joint supervision of instance-pairwise, instance-labeled and class-wise similarities. Specifically, we firstly utilize the instance-pairwise similarity matrix to supervise the learning process of heterogeneous networks and it keeps the pairwise correlation from the perspective of instance-instance. Specially, we design a semantic network to fully exploit the semantic information implicated in labels, which is also used to supervise multi-modal networks on instance-label level. Furthermore, we propose the class-wise hash codes to cooperate with the intrinsic label matrix as the prototypes, and it guides the hash learning and further ensures the precision and compactness of the learned hash codes. In addition, we design different discrete optimization strategies to optimize the class-wise hash codes and unified hash codes, respectively. That avoids the optimization errors and ensures the high-quality of learned hash codes. Experiments on three popular datasets indicate that our method outperforms other state-of-the-art methods in terms of cross-modal retrieval.
Aluminum foam, carbon fiber reinforced plastics (CFRP) and foam-filled structures have been drawn growing attention for their outstanding lightweight and energy absorption capacity; therefore, ...crushing characteristics of a hybrid system involving these components would be of particular interest. In this study, quasi-static compression tests were carried out to experimentally investigate the crushing behaviors of foam-filled aluminum/CFRP hybrid tube subject to transverse loading condition. Based upon the experimental tests and numerical modeling, the interactive effects in between the aluminum foam filler and aluminum/CFRP hybrid tube was explored. It is found that the load carrying and energy absorption capacities of foam-filled hybrid tube were significantly improved in comparison with the summation of net foam filler and empty hybrid tube. The parametric study was carried out for exploring the effects of the aluminum foam density, aluminum tube thickness and ply number of CFRP tube on the crushing behaviors of foam-filled hybrid tube. It is found that both the total energy absorption (EA) of foam-filled hybrid tubes and the EA contributions of the aluminum foam (or aluminum tube, or CFRP tube) were enhanced with increase in density (or thickness, or ply number). The specific energy absorption (SEA) of foam-filled hybrid tubes increased from 3.45 J/g to 9.24 J/g with the foam density changed from 0.23 g cm−3 to 0.70 g cm−3; nevertheless, increase in aluminum tube thickness (or ply number of CFRP tube) has no evident influence on the SEA of foam-filled hybrid tubes. Finally, discrete design optimization was further performed to obtain the best possible foam-filled hybrid configuration for the transverse crushing characteristics. The optimum results showed that the SEA was largely improved by 213% in comparison with the baseline design.
In the component order connectivity problem, we are given a finite undirected graph G = (V,E) and non-negative integers k, ℓ. The goal of this problem is to determine whether there exists a subset S ...⊆ V such that |S| ≤ k and the size of every connected component of the subgraph of G induced by V \ S is at most ℓ. In this paper, we consider the generalization of the component order connectivity problem where the condition on the sizes of connected components is generalized by non-decreasing subadditive functions defined on the subsets of V. We prove that the kernelization techniques for the component order connectivity problem proposed by Xiao can be generalized to our setting.