Powerful Algorithm Research Articles

AdaGuiDE: An adaptive and guided differential evolution for continuous optimization problems

Differential evolution (DE) has been proven as a simple yet powerful meta-heuristic algorithm on tackling continuous optimization problems. Nevertheless most existing DE methods still suffer from certain drawbacks including the use of ineffective mechanisms to adjust mutation strategies and their control parameters that may possibly mislead the search directions, and also the lack of intelligent guidance and reset mechanisms to escape from local optima. Therefore, to enhance the adaptability of DE-based search frameworks and the robustness on optimizing complex problems full of local optima, an adaptive and guided differential evolution (AdaGuiDE) algorithm is proposed. Essentially, the adaptability of the AdaGuiDE search framework is enhanced by three schemes to iteratively refine the search behaviour at two different levels. At the macroscopic level, the AdaGuiDE search framework revises the existing adaptive mechanism for selecting appropriate DE search strategies by counting the actual contributions in terms of solution quality. In addition, the adaption strategy is extended to the microscopic level where a penalty-based guided DE search is employed to guide the search escaping from local optima through temporarily penalizing the local optima and their neighborhood. Furthermore, a systematic boundary revision scheme is introduced to dynamically adjust the search boundary for locating any potential regions of interest during the search. For a rigorous evaluation of the proposed search framework, the AdaGuiDE algorithm is compared against other well-known meta-heuristic approaches on three sets of benchmark functions involving different dimensions in which the AdaGuiDE algorithm attained remarkable results especially on the high-dimensional and complex optimization problems. More importantly, the proposed AdaGuiDE framework shed lights on many possible directions to further enhance the adaptability of the underlying DE-based search strategies in tackling many challenging real-world applications.

A scalable adaptive quadratic kernel method for interpretable epistasis analysis in complex traits.

Our knowledge of the contribution of genetic interactions (epistasis) to variation in human complex traits remains limited, partly due to the lack of efficient, powerful, and interpretable algorithms to detect interactions. Recently proposed approaches for set-based association tests show promise in improving the power to detect epistasis by examining the aggregated effects of multiple variants. Nevertheless, these methods either do not scale to large Biobank data sets or lack interpretability. We propose QuadKAST, a scalable algorithm focused on testing pairwise interaction effects (quadratic effects) within small to medium-sized sets of genetic variants (window size ≤100) on a trait and provide quantified interpretation of these effects. Comprehensive simulations show that QuadKAST is well-calibrated. Additionally, QuadKAST is highly sensitive in detecting loci with epistatic signals and accurate in its estimation of quadratic effects. We applied QuadKAST to 52 quantitative phenotypes measured in ≈300,000 unrelated white British individuals in the UK Biobank to test for quadratic effects within each of 9515 protein-coding genes. We detect 32 trait-gene pairs across 17 traits and 29 genes that demonstrate statistically significant signals of quadratic effects (accounting for the number of genes and traits tested). Across these trait-gene pairs, the proportion of trait variance explained by quadratic effects is comparable to additive effects, with five pairs having a ratio >1. Our method enables the detailed investigation of epistasis on a large scale, offering new insights into its role and importance.

Test and Measure for Partial Mean Dependence Based on Machine Learning Methods

It is of importance to investigate the significance of a subset of covariates W for the response Y given covariates Z in regression modeling. To this end, we propose a significance test for the partial mean independence problem based on machine learning methods and data splitting. The test statistic converges to the standard Chi-squared distribution under the null hypothesis while it converges to a normal distribution under the fixed alternative hypothesis. Power enhancement and algorithm stability are also discussed. If the null hypothesis is rejected, we propose a partial Generalized Measure of Correlation (pGMC) to measure the partial mean dependence of Y given W after controlling for the nonlinear effect of Z. We present the appealing theoretical properties of the pGMC and establish the asymptotic normality of its estimator with the optimal root-N convergence rate. Furthermore, the valid confidence interval for the pGMC is also derived. As an important special case when there are no conditional covariates Z, we introduce a new test of overall significance of covariates for the response in a model-free setting. Numerical studies and real data analysis are also conducted to compare with existing approaches and to demonstrate the validity and flexibility of our proposed procedures. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.

Application on power system economic dispatch of marine predator algorithm improved by asymmetric information exchange

The solution to the economic dispatch (ED) problem for power systems allows the power sector to reduce operating costs. However, the ED problem is a complex nonlinear and nonconvex optimization problem whose solution requires powerful algorithms. We propose a new version of the Marine Predator Algorithm (MPA), called IMPA, for solving complex ED problems. The algorithm introduces an asymmetric information exchange (AIE) mechanism, which not only accelerates to escape of local optima but also enriches the diversity of search. In this work, 12 benchmark functions were used to test the performance of the proposed algorithm IMPA. Then, the IMPA was used to solve the ED engineering problem of power system containing of 6, 13, 40, and complex 140 units. The minimum and average costs searched by IMPA are 1657962.7265$/h and 1657962.7265$/h, and they are much lower than the results of the MPA and NMPA, which means that our proposed improved IMPA improves the performance of MPA for solving the economic dispatch problem of large-scale power systems. The results show that the solutions obtained by IMPA are more competitive than those of MPA and NMPA, which provides an additional solution for cost reduction of the power system.

Self-Adaptive Evolutionary Info Variational Autoencoder

With the advent of increasingly powerful machine learning algorithms and the ability to rapidly obtain accurate aerodynamic performance data, there has been a steady rise in the use of algorithms for automated aerodynamic design optimisation. However, long training times, high-dimensional design spaces and rapid geometry alteration pose barriers to this becoming an efficient and worthwhile process. The variational autoencoder (VAE) is a probabilistic generative model capable of learning a low-dimensional representation of high-dimensional input data. Despite their impressive power, VAEs suffer from several issues, resulting in poor model performance and limiting optimisation capability. Several approaches have been proposed in attempts to fix these issues. This study combines the approaches of loss function modification with evolutionary hyperparameter tuning, introducing a new self-adaptive evolutionary info variational autoencoder (SA-eInfoVAE). The proposed model is validated against previous models on the MNIST handwritten digits dataset, assessing the total model performance. The proposed model is then applied to an aircraft image dataset to assess the applicability and complications involved with complex datasets such as those used for aerodynamic design optimisation. The results obtained on the MNIST dataset show improved inference in conjunction with increased generative and reconstructive performance. This is validated through a thorough comparison against baseline models, including quantitative metrics reconstruction error, loss function calculation and disentanglement percentage. A number of qualitative image plots provide further comparison of the generative and reconstructive performance, as well as the strength of latent encodings. Furthermore, the results on the aircraft image dataset show the proposed model can produce high-quality reconstructions and latent encodings. The analysis suggests, given a high-quality dataset and optimal network structure, the proposed model is capable of outperforming the current VAE models, reducing the training time cost and improving the quality of automated aerodynamic design optimisation.

Optimized task scheduling approach with fault tolerant load balancing using multi-objective cat swarm optimization for multi-cloud environment

Multi-cloud environment enables an organization to access services from more than one cloud service providers the use of multiple cloud computing and it can be treated as single heterogeneous environment. It enables autonomy to run the tasks on private or public cloud based on business or technical requirements. In a multi-cloud platform, load balancing is an essential task to serve the requests from multiple users with different resources effectively. It helps to improve utilization of the cloud resources, throughput, reduce makespan and avoid overload at resources. Load balancing also facilitates the redirection of traffic to resources running in another cloud when a failure occurs in a cloud. Hence, it is more vital to have optimized load balancing methods in multi-cloud infrastructure in order to improve the system performance. This paper presents an optimized fault tolerant load balancing method using multi-objective cat swarm optimization algorithm called MCSOFLB and the results are then compared against other powerful optimization algorithms. The experimental results evidently show that the proposed algorithm ranks first on the whole. The MCSOFLB method produces an average improvement of 31 % makespan, 6 % resource utilization, 12 % cost, 6 % success rate and 32 % average throughput over other benchmark algorithms.

Fabrication of compact substrate integrated waveguide dual band bandpass filter using complementary split-ring resonators for C and X band applications

This article presents a compact dual-band Bandpass Filter(BPF) based substrate integrated waveguide(SIW) used for C and X band applications. Complementary split-ring resonators(CSRR) are loaded into the waveguide surface to obtain more advantages of compactness and good performance. The SIW cavity undergoes initial optimization utilizing the powerful genetic algorithm, renowned for its robustness in solving complex problems, before proceeding to simulation with Ansys HFSS software with loaded CSRRs. The filter displays two passbands: the first has a bandwidth of 2 GHz and resonates at 6.8 GHz, while the second has a bandwidth of 5,55 GHz and resonates at 10.3 GHz. Moreover, with two resonance frequencies, the developed BPF has the benefit of low insertion loss(S21=-1.7dB and -1.9 dB) and improved return loss: (S11=-24dB and -22 dB) respectively. Furthermore, the proposed combination not only allows the implementation of a forward-wave passband propagating below the characteristic cutoff frequency of the waveguide but also exhibits good rejection, demonstrated by the presence of two transmission zeros with rejection of -16 dB and -19 dB, attained at 7.4 GHz and 14.8 GHz respectively. The proposed SIW BPF was designed using a 0.6mm thick FR4 dielectric substrate measuring 13mm × 15mm. The filter is manufactured to validate the concept and the measured results agree well with the simulation. With all these advantages: compact size, high rejection, and good filtering response, the proposed SIW BPF became suitable for microwave applications.

Scalable rapid framework for evaluating network worst robustness with machine learning

Robustness is pivotal for comprehending, designing, optimizing, and rehabilitating networks, with simulation attacks being the prevailing evaluation method. Simulation attacks are often time-consuming or even impractical; however, a more crucial yet persistently overlooked drawback is that any attack strategy merely provides a potential paradigm of disintegration. The key concern is: in the worst-case scenario or facing the most severe attacks, what is the limit of robustness, referred to as “Worst Robustness”, for a given system? Understanding a system’s worst robustness is imperative for grasping its reliability limits, evaluating protective capabilities, and determining associated design and security maintenance costs. To address these challenges, we introduce the concept of Most Destruction Attack (MDA), which is based on the idea of knowledge stacking. MDA is employed to assess the worst robustness of networks, followed by the application of an adapted CNN algorithm to expedite the prediction of worst robustness. We establish the logical validity of MDA and highlight the exceptional performance of the adapted CNN algorithm in predicting the worst robustness across diverse network topologies. This Worst Robustness Evaluation (WRE) framework is scalable, accommodating various attack strategies, whether existing or prospective, and enhancing predictive capabilities with more powerful machine learning algorithms.

Pioneering the plethora of soliton for the (3+1)-dimensional fractional heisenberg ferromagnetic spin chain equation

Abstract In this scholarly article, we investigate the complex structured (3+1)-dimensional Fractional Heisenberg Ferromagnetic Spin Chain equation (FHFSCE) with conformable fractional derivatives. We develop a diverse glut of soliton solutions using an improved version of the G ′ G -expansion method, namely the r + G ′ G -expansion method. The constraints for the existence of these solutions are painstakingly explained. Our findings are clearly communicated through a number of 3D and 2D graphical representations displaying periodic, multiple periodic, kink, and shock soliton solutions. These soliton solutions are expressed in several mathematical function forms, such as hyperbolic, trigonometric, and rational functions. Our findings support the suggested method’s efficacy as a powerful symbolic algorithm for discovering innovative soliton solutions within nonlinear evolution systems.

Woodlog Inventory Optimization using Object Detection and Object Tracking

In the timber industry, keeping track of wood stacks is crucial to avoid stockpiles that are too large or too small ensuring efficient operations. Traditional methods are often time-consuming and inaccurate, leading to inefficiencies. This research proposes a novel solution for timber yard inventory management using cutting-edge technology.The approach utilizes YOLOv8, a powerful object detection algorithm, to identify wood logs in live video feeds. DeepSORT, a tracking algorithm, then follows these identified logs over time. This automation eliminates the need for manual tracking, minimizes errors, and enforces the "First-In-First-Out" (FIFO) principle for efficient inventory use.The study also compares DeepSORT to other tracking algorithms like OC-SORT and ByteTrack to identify the most effective option for this specific application. The results demonstrate that the proposed method significantly improves detection accuracy and reliability, leading to better inventory management.In essence, this research highlights how integrating advanced detection and tracking technologies can revolutionize timber yard operations. By automating processes and ensuring accurate inventory control, these technologies can significantly reduce costs and boost overall efficiency in the timber sector.

The phase diagram of Mercedes Benz model of water using nested sampling algorithm and molecular dynamics simulations

The Mercedes Benz (MB) model of water is quite popular for explaining the properties of water, but the complete phase diagram has not yet been published. In the MB model, water molecules are modelled as two-dimensional Lennard-Jones discs, with three orientation-dependent hydrogen-bonding arms arranged as in the MB logo. Nested Sampling (NS) is a powerful sampling algorithm for the parameter space that directly calculates the partition function. NS can be used for sampling the equilibrium thermodynamics of atomistic systems and allows access to all thermodynamic quantities in absolute terms, including absolute free energies and absolute entropies. Here, NS is used to calculate the phase diagram of the simple two-dimensional Mercedes-Benz (MB) model of water. 32 water particles were used in NS. While this number of particles provides sufficient agreement with the simulations, the results from fewer particles slowly begin to diverge from the result from NS with 32 particles as some of the equilibrium lines begin to shift. By combining the results from NS and molecular dynamics (MD) simulations, different phases were located and identified. In total, 5 different solid phases, 3 liquid phases and one gaseous phase were observed.

A competing risks machine learning study of neutron dose, fractionation, age, and sex effects on mortality in 21,000 mice

This study explores the impact of densely-ionizing radiation on non-cancer and cancer diseases, focusing on dose, fractionation, age, and sex effects. Using historical mortality data from approximately 21,000 mice exposed to fission neutrons, we employed random survival forest (RSF), a powerful machine learning algorithm accommodating nonlinear dependencies and interactions, treating cancer and non-cancer outcomes as competing risks. Unlike traditional parametric models, RSF avoids strict assumptions and captures complex data relationships through decision tree ensembles. SHAP (SHapley Additive exPlanations) values and variable importance scores were employed for interpretation. The findings revealed clear dose–response trends, with cancer being the predominant cause of mortality. SHAP value dose–response shapes differed, showing saturation for cancer hazard at high doses (> 2 Gy) and a more linear pattern at lower doses. Non-cancer responses remained more linear throughout the entire dose range. There was a potential inverse dose rate effect for cancer, while the evidence for non-cancer was less conclusive. Sex and age effects were less pronounced. This investigation, utilizing machine learning, enhances our understanding of the patterns of non-cancer and cancer mortality induced by densely-ionizing radiations, emphasizing the importance of such approaches in radiation research, including space travel and radioprotection.

Federated Many-Task Bayesian Optimization

Bayesian optimization is a powerful surrogate-assisted algorithm for solving expensive black-box optimization problems. While Bayesian optimization was developed for centralized optimization, the availability of massive distributed data has attracted increased interests in exploring federated Bayesian optimization that can use data on multiple clients without leaking the raw data. However, existing federated Bayesian optimization (FBO) approaches assume that either all clients jointly solve the same optimization task, or only one client solves one target optimization task by transferring knowledge from others in a federated way, making them unsuited for many real-world applications. In this paper, we consider FBO for the scenario where multiple related local black-box tasks associated with different clients are jointly optimized by sharing knowledge between tasks without leaking the data privacy. An efficient federated many-task Bayesian optimization framework is proposed to address not independent and identically distributed (non-IID) data while protecting the data privacy in the federated setting. A novel federated knowledge transfer paradigm is developed for dynamic many-task model aggregation according to a dissimilarity matrix. The dissimilarity is measured based on the rank of the predictions and only the hyperparameters in the local Gaussian process models are shared. In addition, a federated ensemble acquisition function is constructed by integrating the predictions of two surrogates using the global and local hyperparameters, respectively, to effectively search for the optimal solution. Experimental results show that our proposed method has reliable performance on both benchmark problems and a real machine learning problem also in the presence of non-IID data.

An Economic Optimization Model of an E-Waste Supply Chain Network: Machine Learned Kinetic Modelling for Sustainable Production

Purpose: E-waste management (EWM) refers to the operation management of discarded electronic devices, a challenge exacerbated due to overindulgent urbanization. The main purpose of this paper is to amalgamate production engineering, statistical methods, mathematical modelling, supported with Machine Learning to develop a dynamic e-waste supply chain model. Method Used: This article presents a multidimensional, cost function-based analysis of the EWM framework structured on three modules including environmental, economic, and social uncertainties in material recovery from an e-waste (MREW) plant, including the production–delivery–utilization process. Each module is ranked using Machine Learning (ML) protocols—Analytical Hierarchical Process (AHP) and combined AHP-Principal Component Analysis (PCA). Findings: This model identifies and probabilistically ranks two key sustainability contributors to the EWM supply chain: energy consumption and carbon dioxide emission. Additionally, the precise time window of 400–600 days from the start of the operation is identified for policy resurrection. Novelty: Ours is a data-intensive model that is founded on sustainable product designing in line with SDG requirements. The combined AHP-PCA consistently outperformed traditional statistical tools, and is the second novelty. Model ratification using real e-waste plant data is the third novelty. Implications: The Machine Learning framework embeds a powerful probabilistic prediction algorithm based on data-based decision making in future e-waste sustained roadmaps.

Evaluation of biochemical algorithms to screen dysbetalipoproteinemia in ε2ε2 and rare APOE variants carriers.

Dysbetalipoproteinemia (DBL) is a combined dyslipidemia associated with an increased risk of atherosclerotic cardiovascular diseases mostly occurring in ε2ε2 subjects and infrequently in subjects with rare APOE variants. Several algorithms have been proposed to screen DBL. In this work, we compared the diagnostic performances of nine algorithms including a new one. Patients were divided into 3 groups according to their APOE genotype: ε2ε2 ("ε2ε2", n=49), carriers of rare variants ("APOEmut", n=20) and non-carriers of ε2ε2 nor APOE rare variant ("controls", n=115). The algorithms compared were those from Fredrickson, Sniderman, Boot, Paquette, De Graaf, Sampson, eSampson, Bea and ours, the "Hospices Civils de Lyon (HCL) algorithm". Our gold standard was the presence of a ε2ε2 genotype or of a rare variant associated with triglycerides (TG) >1.7 mmol/L. A replication in the UK Biobank and a robustness analysis were performed by considering only subjects with both TGand low-density lipoprotein-cholesterol (LDLc) >90th percentile. Total cholesterol (TC)/ApoB and NHDLC/ApoB are the best ratios to suspect DBL. In ε2ε2, according to their likelihood ratios (LR), the most clinically efficient algorithms were the HCL, Sniderman and De Graaf's. In APOEmut, Sniderman's algorithm exhibited the lowest negative LR (0.07) whereas the HCL's exhibited the highest positive LR (29). In both cohorts, the HCL algorithm had the best LR. We proposed a powerful algorithm based onApoB concentration and the routine lipid profile, whichperforms remarkably well in detecting ε2ε2 or APOE variant-related DBL. Additional studies are needed to further evaluate algorithms performances in DBL carriers of infrequent APOE variants.

PMiSLocMF: predicting miRNA subcellular localizations by incorporating multi-source features of miRNAs.

The microRNAs (miRNAs) play crucial roles in several biological processes. It is essential for a deeper insight into their functions and mechanisms by detecting their subcellular localizations. The traditional methods for determining miRNAs subcellular localizations are expensive. The computational methods are alternative ways to quickly predict miRNAs subcellular localizations. Although several computational methods have been proposed in this regard, the incomplete representations of miRNAs in these methods left the room for improvement. In this study, a novel computational method for predicting miRNA subcellular localizations, named PMiSLocMF, was developed. As lots of miRNAs have multiple subcellular localizations, this method was a multi-label classifier. Several properties of miRNA, such as miRNA sequences, miRNA functional similarity, miRNA-disease, miRNA-drug, and miRNA-mRNA associations were adopted for generating informative miRNA features. To this end, powerful algorithms [node2vec and graph attention auto-encoder (GATE)] and one newly designed scheme were adopted to process above properties, producing five feature types. All features were poured into self-attention and fully connected layers to make predictions. The cross-validation results indicated the high performance of PMiSLocMF with accuracy higher than 0.83, average area under the receiver operating characteristic curve (AUC) and area under the precision-recall curve (AUPR) exceeding 0.90 and 0.77, respectively. Such performance was better than all previous methods based on the same dataset. Further tests proved that using all feature types can improve the performance of PMiSLocMF, and GATE and self-attention layer can help enhance the performance. Finally, we deeply analyzed the influence of miRNA associations with diseases, drugs, and mRNAs on PMiSLocMF. The dataset and codes are available at https://github.com/Gu20201017/PMiSLocMF.

An efficient algorithm for resource optimization in TWDM passive optical network using a C-RAN

The traditional base station in C-RAN is divided into three parts: a pool of centralized baseband units (BBUs), a fronthaul network that links the BBUs and remote radio units (RRUs), and RRUs. This paper proposes a novel cooperative algorithm for resource optimization in a time-wavelength division multiplexed (TWDM) passive optical network (PON) incorporating a cloud radio access network (C-RAN). First, a joint collaborative strategy is deployed to optimize cooperative caching and transmission in the wireless and optical domains. Then, the quality of experience (QoE) is improved by bandwidth configuration and caching. Simulation results show that the average throughput of the proposed QoE-aware video cooperative caching and transmission mechanism (QACCTM) algorithm is approximately 30% higher than that of other algorithms. Compared with the relative average residual clutter power (RARCP) and quality-aware wireless edge caching (QAWEC) algorithms, the proposed QACCTM algorithm reduces the access delay by approximately 27.1% and 15.9%, respectively.

RETRACTED: Syah et al. A New Hybrid Algorithm for Multi-Objective Reactive Power Planning via FACTS Devices and Renewable Wind Resources. Sensors2021, 21, 5246.

The Sensors Editorial Office retracts the article, "A New Hybrid Algorithm for Multi-Objective Reactive Power Planning via FACTS Devices and Renewable Wind Resources" [...].

Novel insights into post-marketing AEs associated with leuprorelin: A comprehensive analysis utilizing the FAERS database

PurposeThis research focused on meticulously tracking and identifying adverse reactions associated with leuprorelin, a drug prescribed for conditions such as prostate cancer, endometriosis, uterine fibroids, and early-onset puberty. The main objective was to enhance patient safety and offer informed guidance on the appropriate use of this treatment. MethodsFrom the first quarter of 2004 to the fourth quarter of 2023, a comprehensive analysis was conducted on a significant number of adverse event reports (AERs) from the FDA Adverse Event Reporting System (FAERS) database. Data mining with dismutation analysis was conducted to quantify signals associated with adverse events (AEs) related to leuprorelin, utilizing powerful algorithms such as ROR, PRR, BCPNN, and EBGM. ResultsA total of 102 positive reaction terms (PT) spanning 24 System Organ Classes (SOCs) were identified from an analysis of 60,709 reports associated with leuprorelin use. Notably, several previously unrecognized adverse reactions were uncovered, including Artificial Menopause, Ovarian Adhesion, Follicular Cystitis, Intercepted product preparation error, among others. These findings underscore the importance of exercising additional vigilance regarding the potential adverse effects of leuprorelin, such as Abscess Sterile, Injection site granuloma, Intercepted medication error, and Bulbospinal muscular atrophy congenital. ConclusionsThis research has successfully uncovered new and unforeseen signals associated with adverse drug reactions (ADRs) following leuprorelin administration. The study provides valuable insights into the intricate connection between ADRs and leuprorelin usage. The results underscore the crucial significance of continuous surveillance and meticulous monitoring to promptly identify and manage AEs, ultimately enhancing patient safety and well-being while undergoing leuprorelin therapy.

The possible benefit of artificial intelligence in an organized population-related screening program : Initial results and perspective

Mammography screening programs (MSP) have shown that breast cancer can be detected at an earlier stage enabling less invasive treatment and leading to abetter survival rate. The considerable numbers of interval breast cancer (IBC) and the additional examinations required, the majority of which turn out not to be cancer, are critically assessed. In recent years companies and universities have used machine learning (ML) to develop powerful algorithms that demonstrate astonishing abilities to read mammograms. Can such algorithms be used to improve the quality of MSP? The original screening mammographies of 251 cases with IBC were retrospectively analyzed using the software ProFound AI® (iCAD) and the results were compared (case score, risk score) with acontrol group. The relevant current literature was also studied. The distributions of the case scores and the risk scores were markedly shifted to higher risks compared to the control group, comparable to the results of other studies. Retrospective studies as well as our own data show that artificial intelligence (AI) could change our approach to MSP in the future in the direction of personalized screening and could enable asignificant reduction in the workload of radiologists, fewer additional examinations and areduced number of IBCs; however, the results of prospective studies are needed before implementation.