Adaptive Perturbation Research Articles

An enhanced competitive swarm optimizer with strongly robust sparse operator for large-scale sparse multi-objective optimization problem

In the real world, the decision variables of large-scale sparse multi-objective problems are high-dimensional, and most Pareto optimal solutions are sparse. The balance of the algorithms is difficult to control, so it is challenging to deal with such problems in general. Therefore, An Enhanced Competitive Swarm Optimizer with Strongly Robust Sparse Operator (SR-ECSO) algorithm is proposed. Firstly, the strongly robust sparse functions which accelerate particles in the population better sparsity in decision space, are used in high-dimensional decision variables. Secondly, the diversity of sparse solutions is maintained, and the convergence balance of the algorithm is enhanced by the introduction of an adaptive random perturbation operator. Finally, the state of the particles is updated using a swarm optimizer to improve population competitiveness. To verify the proposed algorithm, we tested eight large-scale sparse benchmark problems, and the decision variables were set in three groups with 100, 500, and 1000 as examples. Experimental results show that the algorithm is promising for solving large-scale sparse optimization problems.

DP-FedEwc: Differentially private federated elastic weight consolidation for model personalization

Federated learning (FL) has become a prevalent paradigm for training a model collaboratively on multiple clients with the coordination of a central server. As traditional FL suffers from client-drift due to data heterogeneity across clients, many personalized FL (PFL) techniques have been proposed. However, the issue of privacy leakage within PFL remains inadequately addressed. When incorporating differential privacy (DP) directly into PFL to provide rigorous privacy guarantees, it amplifies heterogeneity among clients and introduces high variance in uploaded information, significantly compromising model’s utility. In this paper, we propose a novel privacy-preserving PFL framework called Differentially Private Federated Elastic weight consolidation (DP-FedEwc), to achieve effective model personalization for each client under sample-level DP. We focus on a practical setting where the server is honest-but-curious. We first implement a FedEwc algorithm in a communication-efficient manner, and provide privacy guarantees by perturbing models and their parameter importance (PI). We show that FedEwc is robust to DP-introduced heterogeneity caused by noisy models, especially when the model is a deep neural network. Since excessive noise may render PI invalid, we present an Adaptive Parameter importance Perturbation (APP) method to adaptively add Gaussian noise to different coordinates of PI, thereby alleviating the negative effect of DP noise. Moreover, to accurately calibrate the privacy cost resulting from querying heterogeneous data across various clients when computing PI through APP, we adapt a Bayesian Accountant (BA) method to DP-FedEwc. We conduct experiments on standard benchmark datasets, and the experimental results confirm the superiority of DP-FedEwc over DP-PFL baselines.

Simultaneous desulfurization and dearomatization of simulated straight-run diesel with novel green DBN-based ionic liquids

Given the surplus of diesel fuel in China and the extensive utilization of renewable energy sources, there is a growing inclination to eliminate organic sulphur and aromatic hydrocarbons from diesel fuel for its application as high-quality ethylene cracking feedstock. In this study, a series of 1,5-diazabicyclo[4.3.0]non-5-ene (DBN)-based ionic liquids with [DBNH]+ as the cation and imidazole as the anion were designed and synthesized for the simultaneous desulfurization and dearomatization of straight-run diesel fuel. Considering factors such as extractant stability, residue content in oil, and separation performance, [DBNH][Tr], exhibiting excellent performance characteristics, was screened as the optimal extractant. Following three-stage cross-flow extraction with an extractant/oil ratio of 3:1 at 30 °C, complete removals of sulfides (benzothiophene) and polycyclic aromatic hydrocarbons (1-methylnaphthalene) were achieved while saturated alkanes reached up to 95.8 wt%, making it suitable for ethylene cracking applications. The interaction mechanisms between ionic liquids and oil components were revealed by atoms in molecules (AIM) topology analysis, independent gradient model based on Hirshfeld partition (IGMH) analysis, and symmetric adaptive perturbation theory energy decomposition techniques, revealing that dispersion-dominated van der Waals (VDW) forces primarily govern interactions due to π-π stacking.

Photovoltaic model parameters identification using diversity improvement-oriented differential evolution

Fast and accurate parameter identification of the photovoltaic (PV) model is crucial for calculating, controlling, and managing PV generation systems. Numerous meta-heuristic algorithms have been applied to identify unknown parameters due to the multimodal and nonlinear characteristics of the parameter identification problems. Although many of them can obtain satisfactory results, problems such as premature convergence and population stagnation still exist, influencing the optimization performance. A novel variant of Differential Evolution, namely, Diversity Improvement-Oriented Differential Evolution (DIODE), is proposed to mitigate these deficiencies and obtain reliable parameters for PV models. In DIODE, an adaptive perturbation strategy is employed to perturb current individuals to mitigate premature convergence by enhancing population diversity. Secondly, a diversity improvement mechanism is proposed, where information on the covariance matrix and fitness improvement of individuals is used as a diversity indicator to detect stagnant individuals, which are then updated by the intervention strategy. Lastly, a novel parameter adaptation strategy is employed to maintain a sound balance between exploration and exploitation. The proposed DIODE algorithm is applied to parameter identification problems of six PV models, including single, double, and triple diode and three PV module models. In addition, a large test bed containing 72 benchmark functions from CEC2014, CEC2017, and CEC2022 test suites is employed to verify DIODE’s overall performance in terms of optimization accuracy. Experiment results demonstrate that DIODE can secure accurate parameters of PV models and achieve highly competitive performance on benchmark functions.

Adaptive Perturbation for Adversarial Attack.

In recent years, the security of deep learning models achieves more and more attentions with the rapid development of neural networks, which are vulnerable to adversarial examples. Almost all existing gradient-based attack methods use the sign function in the generation to meet the requirement of perturbation budget on L∞ norm. However, we find that the sign function may be improper for generating adversarial examples since it modifies the exact gradient direction. Instead of using the sign function, we propose to directly utilize the exact gradient direction with a scaling factor for generating adversarial perturbations, which improves the attack success rates of adversarial examples even with fewer perturbations. At the same time, we also theoretically prove that this method can achieve better black-box transferability. Moreover, considering that the best scaling factor varies across different images, we propose an adaptive scaling factor generator to seek an appropriate scaling factor for each image, which avoids the computational cost for manually searching the scaling factor. Our method can be integrated with almost all existing gradient-based attack methods to further improve their attack success rates. Extensive experiments on the CIFAR10 and ImageNet datasets show that our method exhibits higher transferability and outperforms the state-of-the-art methods.

Unveiling SAR target recognition networks: Adaptive Perturbation Interpretation for enhanced understanding

Deep neural networks (DNNs) have obtained remarkable achievements in various vision tasks. However, DNNs’ mechanism remains obscure, especially in synthetic aperture radar (SAR) target recognition. To clearly understand DNNs’ mechanism, some interpretation algorithms are proposed to provide a saliency map to visualize DNN’s decisions, which can be categorized into (1) propagation-based methods, (2) activation-based methods, and (3) perturbation-based methods. Unfortunately, each of them has some imperfection in producing saliency maps. The saliency maps produced by propagation-based methods are sensitive to SAR speckles. Activation-based methods have difficulty producing a saliency map where the highlighted region covers the entire feature. Perturbation-based methods tend to forfeit intricate details of the target in saliency map during optimization. To mitigate the above limitations, this paper proposed Adaptive Perturbation Interpretation (API), with two aims: (1) to improve the interpretability of SAR recognition networks and (2) to recover the fine-grained lost during the optimization procedures of perturbation-based methods. Specifically, we first introduce a Multi-information Perturbation Optimization Module (MPOM) to retain the target’s local characteristics while constraining saliency map’s sparsity by configuring regularization terms. Additionally, we further introduce a Feature Compensation Module (FCM) to recover the fine-grained feature lost during the optimization process in MPOM. Extensive experiments are implemented on SAR interpretation benchmark datasets, MSTAR and SARAIRcraft1.0, the results demonstrate the superiority of API to other state-of-the-art methods.

Research on Photovoltaic Maximum Power Point Tracking Control Based on Improved Tuna Swarm Algorithm and Adaptive Perturbation Observation Method

In situations where photovoltaic (PV) systems are exposed to varying light intensities, the conventional maximum power point tracking (MPPT) control algorithm may become trapped in a local optimal state. In order to address this issue, a two-step MPPT control strategy is suggested utilizing an improved tuna swarm optimization (ITSO) algorithm along with an adaptive perturbation and observation (AP&O) technique. For the sake of enhancing population diversity, the ITSO algorithm is initialized by the SPM chaos mapping population. In addition, it also uses the parameters of the spiral feeding strategy of nonlinear processing and the Levy flight strategy adjustment of the weight coefficient to enhance global search ability. In the two-stage MPPT algorithm, the ITSO is applied first to track the vicinity of the global maximum power point (MPP), and then it switches to the AP&O method. The AP&O method’s exceptional local search capability enables the global MPP to be tracked with remarkable speed and precision. To confirm the effectiveness of the suggested algorithm, it is evaluated against fuzzy logic control (FLC), standard tuna swarm optimization (TSO), grey wolf optimization (GWO), particle swarm optimization (PSO), and AP&O. Finally, the proposed MPPT strategy is verified by the MATLAB R2022b and RT-LAB experimental platform. The findings indicate that the suggested method exhibits improved precision and velocity in tracking, efficiently following the global MPP under different shading conditions.

Forecasting a Short-Term Photovoltaic Power Model Based on Improved Snake Optimization, Convolutional Neural Network, and Bidirectional Long Short-Term Memory Network.

The precision of short-term photovoltaic power forecasts is of utmost importance for the planning and operation of the electrical grid system. To enhance the precision of short-term output power prediction in photovoltaic systems, this paper proposes a method integrating K-means clustering: an improved snake optimization algorithm with a convolutional neural network-bidirectional long short-term memory network to predict short-term photovoltaic power. Firstly, K-means clustering is utilized to categorize weather scenarios into three categories: sunny, cloudy, and rainy. The Pearson correlation coefficient method is then utilized to determine the inputs of the model. Secondly, the snake optimization algorithm is improved by introducing Tent chaotic mapping, lens imaging backward learning, and an optimal individual adaptive perturbation strategy to enhance its optimization ability. Then, the multi-strategy improved snake optimization algorithm is employed to optimize the parameters of the convolutional neural network-bidirectional long short-term memory network model, thereby augmenting the predictive precision of the model. Finally, the model established in this paper is utilized to forecast photovoltaic power in diverse weather scenarios. The simulation findings indicate that the regression coefficients of this method can reach 0.99216, 0.95772, and 0.93163 on sunny, cloudy, and rainy days, which has better prediction precision and adaptability under various weather conditions.

An improved white shark optimizer algorithm used to optimize the structural parameters of the oil pad in the hydrostatic bearing

Abstract An improved white shark optimizer (MWSO) algorithm has been proposed. The algorithm adopts an improved tent chaotic mapping strategy to enhance the diversity of the initial population of white sharks, introduces the balance pool strategy of the EO algorithm to improve the convergence speed and accuracy of the algorithm, applies adaptive t-distribution dynamic selection probability perturbation to the global optimal solution, and adjusts the exploration and development ability of the algorithm at different iteration periods. MWSO, WSO, and seven excellent metaheuristic algorithms are tested and compared on 23 classic test functions and the CEC2017 test suite, and two non-parametric tests, a Wilcoxon rank sum test with a significance level of 0.05 and Friedman test, are conducted. The statistical results indicate that the proposed MWSO is significantly superior to other algorithms. In addition, nine algorithms are applied for the first time to optimize the structural parameters of the oil sealing edge of oil pads in response to the issue of the bearing capacity of hydrostatic bearings. This not only further verified the superiority of MWSO, but also provided new ideas for the optimization of hydrostatic bearings.

STP-KDE: A spatiotemporal trajectory protection and publishing method based on kernel density estimation

Promoted by the Internet of Thing era, the widespread use of mobile sensing devices equipped with positioning functions has led to the generation of substantial trajectory data. Mining and analyzing trajectory data has high research value, but poses a risk of user privacy leakage, resulting in fewer publicly available trajectory datasets for research and analysis. Therefore, a trajectory data publishing method that ensures high data utility while protecting user privacy has become a hot topic. In this paper, we propose a spatiotemporal trajectory data protection and publishing method based on kernel density estimation (STP-KDE), which protects the trajectory data and trajectory count values while improving data utility. In the protection process of trajectory data, we designed a kernel density clustering framework that is combined with the differential privacy exponential mechanism. In the protection process of trajectory counts value, the adaptive Laplace noise perturbation mechanism is proposed to differentially protect these counts. Experimental results show that STP-KDE can provide more useful data and stronger privacy protection than existing studies.

Adaptive hybrid optimization algorithm for numerical computing in engineering applications

ABSTRACT A novel hybrid arithmetic optimization algorithm (HAOA) is proposed to address the inherent constraints of traditional numerical computing techniques, including increased computational complexity and excessive reliance on gradient information. First, the Latin hypercube sampling initialization strategy is used to generate higher quality initial candidate solutions. Then, the strategy is used to expand the advantages of leading individuals, boost the local search ability and improve the computational accuracy of the algorithm. Finally, the adaptive t-distribution mutation perturbation strategy is adopted to randomly perturb the position of the current optimal solution, aiming to prevent the algorithm from falling into a local optimal solution. The performance of HAOA is examined in comparison to alternative algorithms using the CEC 2022 data set and Bayesian validation is applied to comprehensively validate the superiority of the HAOA. Numerical experimental results demonstrate that HAOA achieves more accurate extremal and integral results, and has a high solution speed.

Constructing a non-degeneracy nD chaotic map model and counteracting dynamic degradation through adaptive impulsive perturbation

The characteristics of state point trajectory in phase space of a chaotic map, such as ergodicity, aperiodicity, broadband and noise-like, have been proved to be suitable for cryptography. However, some weaknesses, such as small key space, multiple bifurcations and dense windows that exist in some classic 1D and 2D chaotic maps, effect their practical application. To solve these problems, we constructed a non-degeneracy nD chaotic model, and instantiated it to obtain a 3D chaotic map (3D-NDCM), the dynamics analysis results demonstrated that the 3D-NDCM has a sufficiently large chaotic range, its state point trajectories has ergodicity in phase space. To counteract dynamic degradation and extend the aperiodic length of state time sequences, we further constructed a 6D-NDCM based on the model. Experimental analysis results demonstrated that, a state time sequence with longer aperiodicity can be obtained, through both dimension expansion and adaptive impulsive perturbation.

Shore Power System Load Forecasting Model based on IDBO algorithm and PCA-BiLSTM Network

Shore power offers convenient and sustainable power services to ships docking in ports, playing a pivotal role in advancing environmental preservation and the effective utilization of resources. Shore power system loads are susceptible to numerous influencing factors. Therefore, it is crucial to identify a highly accurate load forecasting model to facilitate the judicious allocation of power resources. A short-term shore power system load forecasting model based on the improved dung beetle optimization algorithm (IDBO) and principal component analysis (PCA) in conjunction with the enhanced bidirectional long and short-term memory (BiLSTM) network is proposed in this paper. The IDBO algorithm improves the problem of low convergence accuracy and vulnerability to local optima in DBO. This is achieved by population initialization based on cubic chaotic mapping, amalgamating walrus optimization algorithm, and adaptive t-distribution perturbation strategy. The performance of the BiLSTM, IDBO-BiLSTM, and PCA-IDBO-BiLSTM shore power system load forecasting models are compared through case studies, revealing that the proposed PCA-IDBO-BiLSTM model demonstrates superior predictive capabilities.

Lithium-ion battery SOH prediction based on VMD-PE and improved DBO optimized temporal convolutional network model

Accurately estimating the state of health (SOH) in lithium-ion batteries is crucial for optimizing performance, cost management, enhancing safety, extending lifespan, and promoting environmental sustainability. However, due to factors such as capacity regeneration, battery capacity degradation data exhibits nonlinear and non-stationary characteristics, making it challenging to accurately predict SOH. In order to improve prediction accuracy, this paper proposes a novel SOH prediction model based on VMD-PE-IDBO-TCN. Firstly, the variational mode decomposition-permutation entropy (VMD-PE) method is innovatively introduced to decompose the original battery SOH data and reorganize it into a series of subsequences, reducing the impact of non-stationary components on the prediction results. Moreover, temporal convolutional network (TCN) prediction models are used for each subsequence, and an improved dung beetle optimization (IDBO) algorithm is used to optimize the TCN parameters, where this IDBO algorithm incorporates SPM chaotic mapping, golden sine strategy, and adaptive Gaussian-Cauchy mixed mutation perturbation. The IDBO algorithm rapidly and accurately identifies the optimal parameter combination, significantly enhancing the predictive performance of the TCN model. Finally, the predicted values of each subsequence are superimposed to obtain the final prediction result. Two different lithium-ion battery datasets from NASA and CALCE are used for experimental validation. Experimental results show that the proposed model has higher prediction accuracy, with a maximum RMSE of only 0.0072 and a maximum MAPE of only 0.67 %, and has good adaptability to different types of batteries and batteries operating at high temperatures.

PT-ADP: A personalized privacy-preserving federated learning scheme based on transaction mechanism

Differential privacy (DP) is a widely used technique for enhancing privacy in federated learning (FL) frameworks, whereby noise is added to the datasets or learning parameters to prevent attackers from discovering which client (i.e., the data owner) the sensitive information originated from. More noise implies higher privacy protection but also leads to a decrease in model accuracy. The privacy budget is a key parameter that controls the amount of noise. Most existing methods focus on the same privacy budget, ignoring the various privacy protection requirements of the clients. In this study, we propose PT-ADP, which is a personalized privacy-preserving federated learning scheme. First, a privacy transaction mechanism (PT) is proposed to realize personalized allocation of privacy budget for clients through “quotation-allocation.” Subsequently, an adaptive transmission data perturbation mechanism (ADP) is proposed to save a large amount of privacy budget and improve accuracy under the premise of providing sufficient privacy protection. The security, convergence, efficiency, and specific parameters of the proposed scheme were theoretically analyzed and verified through extensive experiments. PT-ADP can improve model availability while providing the same level of privacy protection compared to other schemes, without increasing complexity and communication overhead.

Adv-Diffusion: Imperceptible Adversarial Face Identity Attack via Latent Diffusion Model

Adversarial attacks involve adding perturbations to the source image to cause misclassification by the target model, which demonstrates the potential of attacking face recognition models. Existing adversarial face image generation methods still can’t achieve satisfactory performance because of low transferability and high detectability. In this paper, we propose a unified framework Adv-Diffusion that can generate imperceptible adversarial identity perturbations in the latent space but not the raw pixel space, which utilizes strong inpainting capabilities of the latent diffusion model to generate realistic adversarial images. Specifically, we propose the identity-sensitive conditioned diffusion generative model to generate semantic perturbations in the surroundings. The designed adaptive strength-based adversarial perturbation algorithm can ensure both attack transferability and stealthiness. Extensive qualitative and quantitative experiments on the public FFHQ and CelebA-HQ datasets prove the proposed method achieves superior performance compared with the state-of-the-art methods without an extra generative model training process. The source code is available at https://github.com/kopper-xdu/Adv-Diffusion.

A Region-Adaptive Local Perturbation-Based Method for Generating Adversarial Examples in Synthetic Aperture Radar Object Detection

In synthetic aperture radar (SAR) imaging, intelligent object detection methods are facing significant challenges in terms of model robustness and application security, which are posed by adversarial examples. The existing adversarial example generation methods for SAR object detection can be divided into two main types: global perturbation attacks and local perturbation attacks. Due to the dynamic changes and irregular spatial distribution of SAR coherent speckle backgrounds, the attack effectiveness of global perturbation attacks is significantly reduced by coherent speckle. In contrast, by focusing on the image objects, local perturbation attacks achieve targeted and effective advantages over global perturbations by minimizing interference from the SAR coherent speckle background. However, the adaptability of conventional local perturbations is limited because they employ a fixed size without considering the diverse sizes and shapes of SAR objects under various conditions. This paper presents a framework for region-adaptive local perturbations (RaLP) specifically designed for SAR object detection tasks. The framework consists of two modules. To address the issue of coherent speckle noise interference in SAR imagery, we develop a local perturbation generator (LPG) module. By filtering the original image, this module reduces the speckle features introduced during perturbation generation. It then superimposes adversarial perturbations in the form of local perturbations on areas of the object with weaker speckles, thereby reducing the mutual interference between coherent speckles and adversarial perturbation. To address the issue of insufficient adaptability in terms of the size variation in local adversarial perturbations, we propose an adaptive perturbation optimizer (APO) module. This optimizer adapts the size of the adversarial perturbations based on the size and shape of the object, effectively solving the problem of adaptive perturbation size and enhancing the universality of the attack. The experimental results show that RaLP reduces the detection accuracy of the YOLOv3 detector by 29.0%, 29.9%, and 32.3% on the SSDD, SAR-Ship, and AIR-SARShip datasets, respectively, and the model-to-model and dataset-to-dataset transferability of RaLP attacks are verified.

Adaptive Fuzzy Finite-Time Singular Perturbation Control for Flexible Joint Manipulators With State Constraints

Adaptive Fuzzy Finite-Time Singular Perturbation Control for Flexible Joint Manipulators With State Constraints

Image-Level Adaptive Adversarial Ranking for Person Re-Identification.

The potential vulnerability of deep neural networks and the complexity of pedestrian images, greatly limits the application of person re-identification techniques in the field of smart security. Current attack methods often focus on generating carefully crafted adversarial samples or only disrupting the metric distances between targets and similar pedestrians. However, both aspects are crucial for evaluating the security of methods adapted for person re-identification tasks. For this reason, we propose an image-level adaptive adversarial ranking method that comprehensively considers two aspects to adapt to changes in pedestrians in the real world and effectively evaluate the robustness of models in adversarial environments. To generate more refined adversarial samples, our image representation enhancement module leverages channel-wise information entropy, assigning varying weights to different channels to produce images with richer information content, along with a generative adversarial network to create adversarial samples. Subsequently, for adaptive perturbation of ranking, the adaptive weight confusion ranking loss is presented to calculate the weights of distances between positive or negative samples and query samples. It endeavors to push positive samples away from query samples and bring negative samples closer, thereby interfering with the ranking of system. Notably, this method requires no additional hyperparameter tuning or extra data training, making it an adaptive attack strategy. Experimental results on large-scale datasets such as Market1501, CUHK03, and DukeMTMC demonstrate the effectiveness of our method in attacking ReID systems.

Feature selection via a multi-swarm salp swarm algorithm

<abstract><p>Feature selection (FS) is a promising pre-processing step before performing most data engineering tasks. The goal of it is to select the optimal feature subset with promising quality from the original high-dimension feature space. The Salp Swarm Algorithm (SSA) has been widely used as the optimizer for FS problems. However, with the increase of dimensionality of original feature sets, the FS problems propose significant challenges for SSA. To solve these issues that SSA is easy to fall into local optimum and have poor convergence performance, we propose a multi-swarm SSA (MSSA) to solve the FS problem. In MSSA, the salp swarm was divided into three sub-swarms, the followers updated their positions according to the optimal leader of the corresponding sub-swarm. The design of multi-swarm and multi-exemplar were beneficial to maintain the swarm diversity. Moreover, the updating models of leaders and followers were modified. The salps learn from their personal historical best positions, which significantly improves the exploration ability of the swarm. In addition, an adaptive perturbation strategy (APS) was proposed to improve the exploitation ability of MSSA. When the swarm stagnates, APS will perform the opposition-based learning with the lens imaging principle and the simulated binary crossover strategy to search for promising solutions. We evaluated the performance of MSSA by comparing it with 14 representative swarm intelligence algorithms on 10 well-known UCI datasets. The experimental results showed that the MSSA can obtain higher convergence accuracy with a smaller feature subset.</p></abstract>