Music dynamically modulates neural activity in memory-related brain regions, enabling more precise retention of details, facilitating easier recall of information, and enhancing working memory capacity. However, most current electroencephalography (EEG)-based memory state recognition approaches neither directly embed music stimuli as inputs nor sufficiently quantify the influence of musical genres. To address this gap, we propose a hierarchical temporal–frequency fusion (HTFF) network that integrates EEG data with musical cues for memory state recognition. Specifically, we propose a music-facilitated EEG temporal–frequency fusion module that progressively fuses low-level perceptual cues into high-level memory representations via cascaded resonant EEG fusion reinforcers (REFRs). These REFR units allocate musical cues into the temporal and frequency domains, thereby enhancing EEG representations through a cross-domain coupling mechanism. We also propose a mixed-constraints mechanism that creates more compact intra-class sample distributions and sharpens decision boundaries for musical genres. The experimental results demonstrate that the HTFF network outperforms state-of-the-art methods in memory state recognition. Moreover, its compatibility with the Internet of Things (IoT) and wearable devices highlights its potential for real-time monitoring and personalized cognitive interventions, paving the way for practical applications in cognitive health.

Existing infrared physical adversarial attack methods struggle to balance strong attack performance and rapid deployment, with fixed iteration steps causing optimization redundancy. We propose an infrared adversarial patch optimization method based on a Gaussian heat diffusion model. By constructing aggregation regularization derived from Fourier's heat conduction law, we precisely guide digital-domain adversarial perturbations to form continuous aggregated shapes, improving physical realizability. We propose the number of connected regions as a compactness metric and, building upon it, design a dual-threshold early-stopping mechanism that further enhances optimization efficiency. Experiments on multiple infrared datasets demonstrate that our method outperforms mainstream approaches in both attack efficacy and optimization efficiency, and exhibits strong cross-model performance across detectors. Remarkably, physical experiments achieve the highest average attack success rate of 91.7% using only a single patch.

Abstract Few‐shot learning (FSL) attempts to learn and optimise the model from a few examples on image classification, which is still threatened by data scarcity. To generate more data as supplements, data augmentation is considered as a powerful and popular technique to enhance the robustness of few‐shot models. However, there are still some weaknesses in applying augmentation methods. For example, all augmented samples have similar semantic information with respect to different augmented transformations, which makes these traditional augmentation methods incapable of learning the property being varied. To address this challenge, we introduce multi‐task learning to learn a primary few‐shot classification task and an auxiliary self‐supervised task, simultaneously. The self‐supervised task can learn transformation property as auxiliary self‐supervision signals to improve the performance of the primary few‐shot classification task. Additionally, we propose a simple, flexible, and effective mechanism for decision fusion to further improve the reliability of the classifier, named model‐agnostic ensemble inference (MAEI). Specifically, the MAEI mechanism can eliminate the influence of outliers for FSL using non‐maximum suppression. Extensive experiment results demonstrate that our method can outperform other state‐of‐the‐art methods by large margins.

Real-time 3-D human action recognition has broad industrial applications, such as surveillance, human–computer interaction, and healthcare monitoring. By relying on complex spatio-temporal local encoding, most existing point cloud sequence networks capture spatio-temporal local structures to recognize 3-D human actions. To simplify the point cloud sequence modeling task, we propose a lightweight and effective point cloud sequence network referred to as SequentialPointNet for real-time 3-D action recognition. Instead of capturing spatio-temporal local structures, SequentialPointNet encodes the temporal evolution of static appearances to recognize human actions. First, we define a novel type of point data, hyperpoint, to better describe the temporally changing human appearances. A theoretical foundation is provided to clarify the information equivalence property for converting point cloud sequences into hyperpoint sequences. Second, the point cloud sequence modeling task is decomposed into a hyperpoint embedding task and a hyperpoint sequence modeling task. Specifically, for hyperpoint embedding, the static point cloud technology is employed to convert point cloud sequences into hyperpoint sequences, which introduces inherent frame-level parallelism; for hyperpoint sequence modeling, a hyperpoint-mixer module is designed as the basic building block to learning the spatio-temporal features of human actions. Extensive experiments on three widely-used 3-D action recognition datasets demonstrate that the proposed SequentialPointNet achieves a competitive classification performance with up to 10× faster than existing approaches.

Transformer-based architecture network has shown excellent performance in the coarse-grained image classification. However, it remains a challenge for the fine-grained image classification task, which needs more significant regional information. As one of the attention mechanisms, transformer pays attention to the most significant region while neglecting other sub-significant regions. To use more regional information, in this letter, we propose a complemental attention multi-feature fusion network (CAMF), which extracts multiple attention features to obtain more effective features. In CAMF, we propose two novel modules: (i) a complemental attention module (CAM) that extracts the most salient attention feature and the complemental attention feature. (ii) a multi-feature fusion module (MFM) that uses different branches to extract multiple regional discriminative features. Furthermore, a new feature similarity loss is proposed to measure the diversity of inter-class features. Experiments were conducted on four public fine-grained classification datasets. Our CAMF achieves 91.2%, 92.8%, 93.3%, 95.3% on CUB-200-2011, Stanford Dogs, FGVC-Aircraft, and Stanford Cars. The ablation study verified that CAM and MFM can focus on more local discriminative regions and improve fine-grained classification performance.

One of the most important and widely faced optimization problems in real applications is the interval multiobjective optimization problems (IMOPs). The state-of-the-art evolutionary algorithms (EAs) for IMOPs (IMOEAs) need a great deal of objective function evaluations to find a final Pareto front with good convergence and even distribution. Further, the final Pareto front is of great uncertainty. In this paper, we incorporate several local searches into an existing IMOEA, and propose a memetic algorithm (MA) to tackle IMOPs. At the start, the existing IMOEA is utilized to explore the entire decision space; then, the increment of the hypervolume is employed to develop an activation strategy for every local search procedure; finally, the local search procedure is conducted by constituting its initial population, whose center is an individual with a small uncertainty and a big contribution to the hypervolume, taking the contribution of an individual to the hypervolume as its fitness function, and performing the conventional genetic operators. The proposed MA is empirically evaluated on ten benchmark IMOPs as well as an uncertain solar desalination optimization problem and compared with three state-of-the-art algorithms with no local search procedure. The experimental results demonstrate the applicability and effectiveness of the proposed MA.

Interval many-objective optimization problems (IMaOPs), involving more than three objectives and at least one subjected to interval uncertainty, are ubiquitous in real-world applications. However, there have been very few effective methods for solving these problems. In this paper, we proposed a set-based genetic algorithm to effectively solve them. The original optimization problem was first transformed into a deterministic bi-objective problem, where new objectives are hyper-volume and imprecision. A set-based Pareto dominance relation was then defined to modify the fast nondominated sorting approach in NSGA-II. Additionally, set-based evolutionary schemes were suggested. Finally, our method was empirically evaluated on 39 benchmark IMaOPs as well as a car cab design problem and compared with two typical methods. The numerical results demonstrated the superiority of our method and indicated that a tradeoff approximate front between convergence and uncertainty can be produced.

Recent works have demonstrated that image descriptors produced by convolutional feature maps provide state-of-the-art performance for image retrieval and classification problems. However, features from a single convolutional layer are not robust enough for shape deformation, scale variation, and heavy occlusion. In this letter, we present a simple and straightforward approach for extracting multiscale (MS) regional maximum activation of convolutions features from different layers of the convolutional neural network. And we also propose aggregating MS features into a single vector by a parameter-free hedge method for image retrieval. Extensive experimental results on three challenging benchmark datasets indicate that the proposed method achieved outstanding performance against state-of-the-art methods.

Multi-objective optimization problems with interval parameters (IMOPs) are ubiquitous in real-world applications. The existing evolutionary algorithms for IMOPs (IMOEAs) require a large amount of function evaluations to generate an approximate Pareto front which is well converged and evenly distributed, and the generated front has uncertainties to a large extent. In this paper, a local search is embedded into an existing IMOEA, and a memetic algorithm for IMOPs is developed. The existing IMOEA is first employed to search the entire search space, and then the rate of changes of hypervolume is utilized to design an activation mechanism to specify when to conduct the local search. Finally, an initial population of the local search is created by taking the individuals with a large contribution to hypervolume and a small imprecision as the center, and the local search is implemented by taking the contribution to hypervolume as its fitness function. The proposed algorithm is applied to six benchmark IMOPs and an uncertain optimization problem of solar desalination, and compared with a typical IMOEA without the local search. The empirical results indicate the effectiveness of the proposed algorithm.

Interval programming problems are ubiquitous in real-world situations. There exist a variety of theories and methods for handling them; the existing methods, however, have adopted various dominance criteria to distinguish solutions, and these criteria are always subjective. Different dominance criteria will produce different optimal solution(s), and subjective criteria make users, especially for those who are not familiar with interval arithmetic, difficult to choose, which restricts their widespread applications. In this study, the idea of ensemble dominance on intervals for tackling these problems is proposed. Dominance criteria on intervals are first defined, and their correlations are depicted by equivalent, inclusive and non-included relations; then, a reduction scheme is derived by investigating the influence of different criteria on the ordering of solutions, and a novel ensemble dominance relation on intervals is defined to rationally and equally evaluate the quality of a solution; furthermore, the complexity of the proposed method is analyzed; finally, empirical results indicate the effectiveness of the proposed method.

Visual object tracking is a challenging task due to object appearance changes caused by shape deformation, heavy occlusion, background clutters, illumination variation, and camera motion. In this letter, we propose a novel robust algorithm which decomposes the task of tracking into translation and scale estimation. We estimate the translation by using five correlation filters with hierarchical convolutional features which produced multilevel correlation response maps to collaboratively infer the target location. We also calculate the scale variation by another correlation filter with histogram of oriented gradient features at the same time. Extensive experimental results on a large-scale 50 challenging benchmark dataset show that the proposed algorithm achieved outstanding performance against state-of-the-art methods.

Robust scale calculation is a challenging problem in visual tracking. Most existing trackers fail to handle large scale variations in complex videos. To address this issue, we propose a robust and efficient scale calculation method in tracking-by-detection framework, which divides the target into four patches and computes the scale factor by finding the maximum response position of each patch via color attributes kernelized correlation filter. In particular, we employ the weighting coefficients to remove the abnormal matching points and transform the desired training output of the conventional classifier to solve the location ambiguity problem. Experiments are performed on several challenging color sequences with scale variations in the recent benchmark evaluation. And the results show that our method outperforms state-of-the-art tracking methods while operating in real-time.

RDF ontology can represent semantic information in semantic Web. Ontologies are often too large to be used in a single application. Extracting sub-ontology from large-scale ontology can solve this problem. RDF(S) ontologies are abstracted as graph models. According to the RDFS inference rules, the closure of RDFS ontology graph model can be constructed. To decrease the time of generating closure, a parallel closure algorithm is presented. We also present an algorithm based on the graph theory to extract a sub-graph with the concepts in the domain term lexicon as nodes. This sub-graph is regarded as the required domain ontology. The method is applicable to RDF(S) domain ontology extraction. Applying the method, domain ontology can be built fast and effectively.