Search Results (8,693)

To address the challenges of black-box video adversarial attacks, such as excessive query times and suboptimal attack performance due to the lack of result feedback during the attack process, we propose a reinforcement learning-based sparse adversarial attack method called RLVS. This approach leverages reinforcement learning to identify key frames for efficient gradient estimation, significantly reducing the number of queries. First, a self-attention network is integrated into the agent policy network to enable more precise selection of key frames. Second, designed reward functions allow the agent to continuously adapt to the sparse key frames by querying the black-box threat model and receiving feedback on attack outcomes. Lastly, gradient estimation is applied solely to the selected key frames, estimating only the gradient sign rather than the full gradient, further enhancing attack efficiency. We conducted experiments on two video recognition models using three popular action datasets. The experimental results demonstrate that our method outperforms other black-box video attack methods in terms of attack efficiency and effectiveness, achieving higher fooling rates with fewer queries and minimal perturbations. Full article

Bisphenol S (BPS) is a typical endocrine disruptor associated with obesity. To observe BPS effects on lipid metabolism in HepG2 and SK-Hep-1 human HCC cells, a CCK-8 assay was used to assess cell proliferation in response to BPS, and the optimal concentration of BPS was selected. Biochemical indices such as triglyceride (TG) and total cholesterol (T-CHO), and oxidative stress indices such as malondialdehyde (MDA) and catalase (CAT) were measured. ROS and MDA levels were significantly increased after BPS treatment for 24 h and 48 h (p < 0.05), indicating an oxidative stress response. Alanine aminotransferase (ALT), T-CHO, and low-density lipoprotein cholesterol (LDL-C) levels also increased significantly after 24 or 48 h BPS treatments (p < 0.05). RT-PCR and Western blot analyses detected mRNA or protein expression levels of peroxisome proliferator-activated receptor α (PPARα) and sterol regulatory element-binding protein 1c (SREBP1C). The results indicated that BPS could inhibit the mRNA expression of PPARα and carnitine palmitoyl transferase 1B (CPT1B), reduce lipid metabolism, promote mRNA or protein expression of SREBP1C and fatty acid synthase (FASN), and increase lipid synthesis. Increased lipid droplets were observed using morphological Oil Red O staining. Our study demonstrates that BPS may cause lipid accumulation by increasing oxidative stress and perturbing cellular lipid metabolism. Full article

In this paper, the influence of the viscoelasticity of boundary constraints on the forced vibration of the nonlinear forced resonance of a fluid-conveying layered pipe under an external forced excitation is studied. The pipe lays on viscoelastic foundations and is simply supported at both ends, and one end is subject to a viscoelastic boundary constraint. The Kelvin–Voight model was employed to describe the viscoelasticity provided by the foundation and boundary constraint. Hamilton’s variational principle was used to obtain the governing equations, during which geometric nonlinear factors including curvature nonlinearity and inertia nonlinearity were considered. By employing a perturbation-incremental harmonic balance method (IHBM), amplitude–frequency bifurcation diagrams of the pipe were obtained. The results show that the viscoelastic constraints from the boundary and foundation have significant influence on the linear and nonlinear dynamic behavior of the pipe system. Full article

This manuscript aims to increase the utilization of solar energy, which is both environmentally friendly and easily accessible, to satisfy the energy needs of developing countries. In order to achieve this goal, maximum power generation should be provided from photovoltaic panels. Several maximum power point tracking (MPPT) methods are utilized for maximum power generation in photovoltaic panel systems under different weather conditions. In this paper, a novel intelligent hybrid fuzzy logic-controlled maximum power point tracking algorithm founded on the perturb and observe (PO) algorithm is presented. The proposed fuzzy logic controller algorithm and the incremental conductivity maximum power point tracking algorithm were simulated in a MATLAB(2018b version)/Simulink environment and evaluated by comparing the results. Four Sharp ND-F4Q295 solar panels, two in series and two in parallel, were used for the simulation. In this study, the voltage ripple of the proposed hybrid method was measured at 1% compared to the classical incremental conductivity method, while it was 8.6% in the IncCon method. Similarly, the current ripple was 1.08% in the proposed hybrid FLC method, while the current ripple was 9.27% in the IncCon method. It is observed that the proposed smart method stabilizes the system voltage faster, at 25 ms, in the event of sudden weather changes. Full article

Several indicators have been recently proposed for the measurement of various characteristics of the tuples of a dataset—particularly the so-called skyline tuples, i.e., those that are not dominated by other tuples. Numeric indicators are very important as they may, e.g., provide an additional criterion to be used to rank skyline tuples and focus on a subset thereof. We focus on an indicator of robustness that may be measured for any skyline tuple t: the grid resistance, i.e., how large-value perturbations can be tolerated for t to remain non-dominated (and thus in the skyline). The computation of this indicator typically involves one or more rounds of computation of the skyline itself or, at least, of dominance relationships. Building on recent advances in partitioning strategies allowing the parallel computation of skylines, we discuss how these strategies can be adapted to the computation of the indicator. Full article

Overexposure of humans to heavy metals and essential metals poses a significant risk for the development of neurological and neurodevelopmental disorders. The mechanisms through which these metals exert their effects include the generation of reactive oxygen species, mitochondrial dysfunction, activation of inflammatory pathways, and disruption of cellular signaling. The function of glial cells in brain development and in the maintenance of homeostasis cannot be overlooked. The glial cells are particularly susceptible to metal-induced neurotoxicity. Accumulation of metals in the brain promotes microglial activation, triggering inflammatory responses that can coincide with other mechanisms of neurotoxicity, inducing alteration in synaptic transmission, cognitive deficit, and neuronal damage. In this review, we highlighted the role of glial dysfunction in some selected neurodegenerative diseases and neurodevelopmental disorders. We further dive into how exposure to metals such as nickel, manganese, methyl mercury, cadmium, iron, arsenic, and lead affect the functions of the microglia, astrocytes, and oligodendrocytes and the mechanisms through which they exert the effects on the brain in relation to some selected neurodegenerative diseases and neurodevelopmental disorders. Potential therapeutic interventions such as the use of new and improved chelating agents and antioxidant therapies might be a significant approach to alleviating these metal-induced glial perturbations. Full article

Determining the values of various properties for new bio-inks for 3D printing is a very important task in the design of new materials. For this purpose, a large number of experimental works have been consulted, and a database with more than 1200 bioprinting tests has been created. These tests cover different combinations of conditions in terms of print pressure, temperature, and needle values, for example. These data are difficult to deal with in terms of determining combinations of conditions to optimize the tests and analyze new options. The best model demonstrated a specificity (Sp) of 88.4% and a sensitivity (Sn) of 86.2% in the training series while achieving an Sp of 85.9% and an Sn of 80.3% in the external validation series. This model utilizes operators based on perturbation theory to analyze the complexity of the data. For comparative purposes, neural networks have been used, and very similar results have been obtained. The developed tool could easily be applied to predict the properties of bioprinting assays in silico. These findings could significantly improve the efficiency and accuracy of predictive models in bioprinting without resorting to trial-and-error tests, thereby saving time and funds. Ultimately, this tool may help pave the way for advances in personalized medicine and tissue engineering. Full article

Nanoporous membranes are heterogeneous structures, with heterogeneity manifesting at the microscale. In examining particle transport through such media, it has been observed that this transport deviates from classical diffusion, as described by Fick’s second law. Moreover, the classical model is physically unsustainable, as it is non-causal and predicts an infinite speed of concentration perturbation propagation through a substantial medium. In this work, we have derived two causal models as extensions of Fick’s second law, where causality is linked to the effects of inertial memory in the nanoporous membrane. The results of the derived models have been compared with each other and with those obtained from the classical model. It has been demonstrated that both causal models, one with exponentially fading inertial memory and the other with power-law fading memory, predict that the concentration perturbation propagates as a damped wave, leading to an increased time required for the cumulative amount of molecules passing through the membrane to reach a steady state compared to the classical model. The power-law fading memory model predicts a longer time required to achieve a stationary state. These findings have significant implications for understanding cell physiology, developing drug delivery systems, and designing nanoporous membranes for various applications. Full article

Deep neural networks have been shown to produce incorrect predictions when imperceptible perturbations are introduced into the clean input. This phenomenon has garnered significant attention and extensive research in 2D images. However, related work on point clouds is still in its infancy. Current methods suffer from issues such as generated point outliers and poor attack generalization. Consequently, it is not feasible to rely solely on overall or geometry-aware attacks to generate adversarial samples. In this paper, we integrate adversarial transfer networks with the geometry-aware method to introduce adversarial loss into the attack target. A state-of-the-art autoencoder is employed, and sensitivity maps are utilized. We use the autoencoder to generate a sufficiently deceptive mask that covers the original input, adjusting the critical subset through a geometry-aware trick to distort the point cloud gradient. Our proposed approach is quantitatively evaluated in terms of the attack success rate (ASR), imperceptibility, and transferability. Compared to other baselines on ModelNet40, our method demonstrates an approximately 38% improvement in ASR for black-box transferability query attacks, with an average query count of around 7.84. Comprehensive experimental results confirm the superiority of our method. Full article

Increasing demand for nutritious, safe, and healthy food, including the need to preserve biodiversity and other resources, signifies a prodigious challenge for agriculture, which is already at risk from climate change. Diverse and healthy plant-based diets may significantly reduce food insecurity, malnutrition, diet-related diseases, and other health-related issues. More attention to agricultural systems diversity is mandatory to improve the economic, environmental, ecological, and social sustainability of food production in developing countries. In this context, neglected legume production could significantly provide nutritional and healthy benefits for people while adhering to sustainability principles. However, the contribution of neglected legumes to food and nutrition security is still limited due to socio-economic challenges faced by farmers that contribute to the underutilization of neglected legumes, leading to overreliance on a few legumes with poor resilience to climatic perturbations, thus posing a risk to sustainable food production. While major legumes offer higher economic returns and more developed value chains, they also contribute to environmental degradation and resource depletion. Neglected legumes, on the other hand, provide ecosystem services, promote biodiversity, and offer climate resilience but face economic challenges due to limited market demand and underdeveloped value chains. Consequently, food nutritional insecurity and human health concerns remain prevalent, especially in developing countries. There is an urgent need to promote neglected legumes in agricultural systems through policy change implementation, genetic improvement, and development, fostering international cooperation to share knowledge, technologies, and best practices in the production and utilization of neglected legumes. This review comprehensively explores the utility of neglected legumes for food, nutritional security, and human health. It identifies knowledge gaps that should be prioritized as part of research strategies for sustainable future food systems in sub-Saharan Africa. Full article

The Dung Beetle Optimization Algorithm (DBO) is characterized by its great convergence accuracy and quick convergence speed. However, like other swarm intelligent optimization algorithms, it also has the disadvantages of having an unbalanced ability to explore the world and to use local resources, as well as being prone to settling into local optimal search in the latter stages of optimization. In order to address these issues, this research suggests a multi-strategy fusion dung beetle optimization method (MSFDBO). To enhance the quality of the first solution, the refractive reverse learning technique expands the algorithm search space in the first stage. The algorithm’s accuracy is increased by adding an adaptive curve to control the dung beetle population size and prevent it from reaching a local optimum. In order to improve and balance local exploitation and global exploration, respectively, a triangle wandering strategy and a fusion subtractive averaging optimizer were later added to Rolling Dung Beetle and Breeding Dung Beetle. Individual beetles will congregate at the current optimal position, which is near the optimal value, during the last optimization stage of the MSFDBO; however, the current optimal value could not be the global optimal value. Thus, to variationally perturb the global optimal solution (so that it leaps out of the local optimal solution in the final optimization stage of the MSFDBO) and to enhance algorithmic performance (generally and specifically, in the effect of optimizing the search), an adaptive Gaussian–Cauchy hybrid variational perturbation factor is introduced. Using the CEC2017 benchmark function, the MSFDBO’s performance is verified by comparing it to seven different intelligence optimization algorithms. The MSFDBO ranks first in terms of average performance. The MSFDBO can lower the labor and production expenses associated with welding beam and reducer design after testing two engineering application challenges. When it comes to lowering manufacturing costs and overall weight, the MSFDBO outperforms other swarm intelligence optimization methods. Full article

This paper presents the design, simulation, and experimental evaluation of a low-cost, fixed-step MPPT algorithm based on the incremental conductance technique for operation in a low-power photovoltaic (PV) system with a full-bridge DC-AC converter. The performance of the MPPT algorithm was improved by selecting an appropriate fixed perturbation step size and frequency, ensuring efficient power tracking. The implementation was further optimized by restructuring the conventional algorithm and adapting the DC-AC converter control parameters, which enhanced overall performance and optimized coupling for AC loads. The simulation was performed in Simulink/Matlab with a 560 Wp PV system and a resistive load, under variable irradiation conditions. The perturbation step size was set to 1%, and the perturbation frequency ranged between 2 Hz and 15 Hz, with the converter output at 60 Hz. Experimentally, it was validated at an irradiance of 1000 W/m² and an ambient temperature of 45 °C. The algorithm achieved simulation efficiencies of up to 98.93% and an average experimental efficiency of 96.76%. The response time improved by 86% with a perturbation frequency of 15 Hz. This developed MPPT algorithm demonstrates its reliability, accuracy, and feasibility for implementation. Full article

The scarcity of high-quality labeled data poses a challenge to the application of deep learning (DL) in landslide identification from remote sensing (RS) images. Semi-supervised learning (SSL) has emerged as a promising approach to address the issue of low accuracy caused by the limited availability of high-quality labels. Nevertheless, the application of SSL approaches developed for natural images to landslide identification encounters several challenges. This study focuses on two specific challenges: inadequate information extraction from limited unlabeled RS landslide images and the generation of low-quality pseudo-labels. To tackle these challenges, we propose a novel and generic DL framework called hybrid perturbation mean match (HPM-Match). The framework combines dual-branch input perturbation (DIP) and independent triple-stream perturbation (ITP) techniques to enhance model accuracy with limited labels. The DIP generation approach is designed to maximize the utilization of manually pre-defined perturbation spaces while minimizing the introduction of erroneous information during the weak-to-strong consistency learning (WSCL) process. Moreover, the ITP structure unifies input, feature, and model perturbations, thereby broadening the perturbation space and enabling knowledge extraction from unlabeled landslide images across various perspectives. Experimental results demonstrate that HPM-Match has substantial improvements in IoU, with maximum increases of 26.68%, 7.05%, and 12.96% over supervised learning across three datasets with the same label ratio and reduces the number of labels by up to about 70%. Furthermore, HPM-Match strikes a better balance between precision and recall, identifying more landslides than other state-of-the-art (SOTA) SSL approaches. Full article

We propose the adaptive t-distribution spiral search Dung Beetle Optimization (TSDBO) Algorithm to address the limitations of the vanilla Dung Beetle Optimization Algorithm (DBO), such as vulnerability to local optima, weak convergence speed, and poor convergence accuracy. Specifically, we introduced an improved Tent chaotic mapping-based population initialization method to enhance the distribution quality of the initial population in the search space. Additionally, we employed a dynamic spiral search strategy during the reproduction phase and an adaptive t-distribution perturbation strategy during the foraging phase to enhance global search efficiency and the capability of escaping local optima. Experimental results demonstrate that TSDBO exhibits significant improvements in all aspects compared to other modified algorithms across 12 benchmark tests. Furthermore, we validated the practicality and reliability of TSDBO in robotic path planning applications, where it shortened the shortest path by 5.5–7.2% on a 10 × 10 grid and by 11.9–14.6% on a 20 × 20 grid. Full article

Deep learning models have been widely applied to synthetic aperture radar (SAR) target recognition, offering end-to-end feature extraction that significantly enhances recognition performance. However, recent studies show that optical image recognition models are widely vulnerable to adversarial examples, which fool the models by adding imperceptible perturbation to the input. Although the targeted adversarial attack (TAA) has been realized in the white box setup with full access to the SAR model’s knowledge, it is less practical in real-world scenarios where white box access to the target model is not allowed. To the best of our knowledge, our work is the first to explore transferable TAA on SAR models. Since contrastive learning (CL) is commonly applied to enhance a model’s generalization, we utilize it to improve the generalization of adversarial examples generated on a source model to unseen target models in the black box scenario. Thus, we propose the contrastive learning-based targeted adversarial attack, termed CL-TAA. Extensive experiments demonstrated that our proposed CL-TAA can significantly improve the transferability of adversarial examples to fool the SAR models in the black box scenario. Full article