Speed Of Algorithm Research Articles

A novel hybridization approach to improve the critical distance clustering algorithm: Balancing speed and quality

Clustering is a prominent research area, with numerous studies and the development of hundreds of algorithms over the years. However, a fundamental challenge in clustering research is the trade-off between algorithm speed and clustering quality. Existing algorithms tend to prioritize either fast execution with compromised clustering quality or slower performance with superior clustering results. In this study, we propose a novel CDC-2 algorithm, an improved version of the Critical Distance Clustering (CDC) algorithm, to address this challenge. Inspired by the concepts of hybridization in biology and the division of labor in the economic system, we present a new hybridization strategy. Our approach integrates the connectivity and coherence aspects of the K-means and CDC-2 algorithms, respectively, allowing us to combine speed and quality in a single algorithm. This approach is referred to as the CDC++ algorithm, and it is characterized as a hybrid that combines elements from two algorithms, K-means and CDC-2, in order to leverage their strengths while mitigating their weaknesses. Moreover, the structure and mechanism of the CDC++ algorithm led to the introduction of a new concept called “object autoencoder.” Unlike traditional feature reduction methods, this concept focuses on object reduction, representing a significant advancement in clustering techniques. To validate our approach, we conducted experimental studies on thirteen synthetic and five real datasets. Comparative analysis with four well-known algorithms demonstrates that our proposed development and hybridization enable efficient processing of large-scale and high-dimensional datasets without compromising clustering quality.

Intelligent reflecting surface‐assisted UAV inspection system based on transfer learning

AbstractIntelligent reflective surface (IRS) provides an effective solution for reconfiguring air‐to‐ground wireless channels, and intelligent agents based on reinforcement learning can dynamically adjust the reflection coefficient of IRS to adapt to changing channels. However, most exiting IRS configuration schemes based on reinforcement learning require long training time and are difficult to be industrially deployed. This paper, proposes a model‐free IRS control scheme based on reinforcement learning and adopts transfer learning to accelerate the training process. A knowledge base of the source tasks has been constructed for transfer learning, allowing accumulation of experience from different source tasks. To mitigate potential negative effects of transfer learning, quantitative analysis of task similarity through unmanned aerial vehicle (UAV) flight path is conducted. After identifying the most similar source task to the target task, parameters of the source task model are used as the initial values for the target task model to accelerate the convergence process of reinforcement learning. Simulation results demonstrate that the proposed method can increase the convergence speed of the traditional DDQN algorithm by up to 60%.

Stabilized quantum-enhanced SIEM architecture and speed-up through Hoeffding tree algorithms enable quantum cybersecurity analytics in botnet detection.

For the first time, we enable the execution of hybrid quantum machine learning (HQML) methods on real quantum computers with 100 data samples and real-device-based simulations with 5000 data samples, thereby outperforming the current state of research of Suryotrisongko and Musashi from 2022 who were dealing with 1000 data samples and quantum simulators (pure software-based emulators) only. Additionally, we beat their reported accuracy of 76.8% by an average accuracy of 91.2%, all within a total execution time of 1687 s. We achieve this significant progress through two-step strategy: Firstly, we establish a stable quantum architecture that enables us to execute HQML algorithms on real quantum devices. Secondly, we introduce new hybrid quantum binary classifiers (HQBCs) based on Hoeffding decision tree algorithms. These algorithms speed up the process via batch-wise execution, reducing the number of shots required on real quantum devices compared to conventional loop-based optimizers. Their incremental nature serves the purpose of online large-scale data streaming for domain generation algorithm (DGA) botnet detection, and allows us to apply HQML to the field of cybersecurity analytics. We conduct our experiments using the Qiskit library with the Aer quantum simulator, and on three different real quantum devices from Azure Quantum: IonQ, Rigetti, and Quantinuum. This is the first time these tools are combined in this manner.

Query Optimization in Distributed Database Based on Improved Artificial Bee Colony Algorithm

Query optimization is one of the key factors affecting the performance of database systems that aim to enact the query execution plan with minimum cost. Particularly in distributed database systems, due to the multiple copies of the data that are stored in different data nodes, resulting in the dramatic increase in the feasible query execution plans for a query statement. Because of the increasing volume of stored data, the cluster size of distributed databases also increases, resulting in poor performance of current query optimization algorithms. In this case, a dynamic perturbation-based artificial bee colony algorithm is proposed to solve the query optimization problem in distributed database systems. The improved artificial bee colony algorithm improves the global search capability by combining the selection, crossover, and mutation operators of the genetic algorithm to overcome the problem of falling into the local optimal solution easily. At the same time, the dynamic perturbation factor is introduced so that the algorithm parameters can be dynamically varied along with the process of iteration as well as the convergence degree of the whole population to improve the convergence efficiency of the algorithm. Finally, comparative experiments conducted to assess the average execution cost of Top-k query plans generated by the algorithms and the convergence speed of algorithms under the conditions of query statements in six different dimension sets. The results demonstrate that the Top-k query plans generated by the proposed method have a lower execution cost and a faster convergence speed, which can effectively improve the query efficiency. However, this method requires more execution time.

Identification algorithm for power quality disturbance of inspection robot in converter station during operation

AbstractAs power quality disturbances become increasingly complex, it is imperative that the speed and accuracy of the inspection robot at the converter station be improved. For this purpose, this study designs a power quality disturbance feature extraction method based on the fast‐adaptive S‐transform. This method preserves the main feature information and eliminates redundant calculations on the basis of adaptive transform. On this basis, a power quality disturbance identification model built on a multi label lightweight gradient elevator is constructed. In the experimental results, compared to the generalized S‐transform, adaptive S‐transform, and S‐transform, the total extraction time of the proposed method was reduced by 96.09%, 91.56%, and 91.22%, respectively. The average accuracy of extracting features for a single disturbance was 99.56%, while for complex disturbance, it was 98.24%, both of which outperformed the comparison algorithms. It is verified that the proposed method can improve the extraction accuracy of power quality disturbance signal. In the recognition of a single disturbance signal, the constructed model exhibited a high accuracy of over 99%. In recognizing composite disturbance signals, the model demonstrated high accuracy and strong stability. Its effectiveness has been confirmed through experiments. The paper aims to enhance the speed and accuracy of power quality disturbance recognition algorithms. This will assist inspection robots working in converter stations, and ensure stable and safe operation of the power grid.

CasOmniMVS: Cascade Omnidirectional Depth Estimation with Dynamic Spherical Sweeping

Estimating 360∘ depth from multiple cameras has been a challenging problem. However, existing methods often adopt a fixed-step spherical sweeping approach with densely sampled spheres and use numerous 3D convolutions in networks, which limits the speed of algorithms in practice. Additionally, obtaining high-precision depth maps of real scenes poses a challenge for the existing algorithms. In this paper, we design a cascade architecture using a dynamic spherical sweeping method that progressively refines the depth estimation from coarse to fine over multiple stages. The proposed method adaptively adjusts sweeping intervals and ranges based on the predicted depth and the uncertainty from the previous stage, resulting in a more efficient cost aggregation performance. The experimental results demonstrated that our method achieved state-of-the-art accuracy with reduced GPU memory usage and time consumption compared to the other methods. Furthermore, we illustrate that our method achieved satisfactory performance on real-world data, despite being trained on synthetic data, indicating its generalization potential and practical applicability.

AlphaCRV: a pipeline for identifying accurate binder topologies in mass-modeling with AlphaFold.

The speed and accuracy of deep learning-based structure prediction algorithms make it now possible to perform in silico "pull-downs" to identify protein-protein interactions on a proteome-wide scale. However, on such a large scale, existing scoring algorithms are often insufficient to discriminate biologically relevant interactions from false positives. Here, we introduce AlphaCRV, a Python package that helps identify correct interactors in a one-against-many AlphaFold screen by clustering, ranking, and visualizing conserved binding topologies, based on protein sequence and fold. AlphaCRV is a Python package for Linux, freely available at https://github.com/strubelab/AlphaCRV.

A new pilot assignment scheme for mitigating pilot contamination in uplink massive multi-input–multi-output (MIMO) systems

Pilot contamination is a serious issue in massive multi–input–multi–output systems which significantly degrades system performance. In this paper, we investigate a new pilot assignment scheme by integrating two-dimensional genetic algorithm with Tabu-Search algorithm (TS) to mitigate the pilot contamination problem. Firstly, we design a two-dimensional genetic algorithm equipped with elitism strategy as a principal algorithm for solving the pilot assignment problem; then, aiming to enhance the convergence speed of the genetic algorithm to the ideal optimal solution, we integrate TS with the genetic algorithm. This integrated pilot assignment scheme, henceforth designated as GATS-PA, is found to be powerful in mitigating the pilot contamination problem. Numerical simulation results verify that the proposed pilot assignment scheme is very close to the ideal optimal solution with few numbers of iterations and outperforms existing methods in terms of enhancing the average uplink rate per user over a wide range of simulation parameters.

Exploring the Teaching Mode of English for Specialized Purposes (ESP) in Colleges and Universities under the PDCA Model

Abstract In this paper, ant colony algorithm and genetic algorithm are combined and applied to the teaching mode of English for Special Purposes (ESP) in colleges and universities, and the optimization of the teaching mode is realized through the PDCA model. The iteration speed of ant colony algorithm is better than that of genetic algorithm in the early stage. In comparison, the iteration speed of genetic algorithm is better in the later stage. By fusing these two algorithms, their respective advantages can be maximized. The study takes two computer science classes as an example, first conducts a needs analysis of the ESP course, and finds that the ratings in English foundation, learning materials, learning mode, and quiz mode are all higher than 5.5. By comparing the final exam scores of the experimental and control classes, it was found that the average score of the practical class was 3.5 points higher than that of the control class, which indicated that the ESP teaching of the PDCA cycle effectively improved the students’ academic performance. The average score of students’ recognition of the ESP teaching model of PDCA cycle was 4.252, while the ratings in satisfaction, participation in classroom interactions, and learning effect were all over 4.1. The analysis of learning motivation shows that “for better employment”, “for promotion”, and “to pass the exam” are the top three motivations for students to choose to learn English for computer science majors, with scores of 175, 174, and 170, respectively. This study effectively demonstrates the potential of combining ACO and Genetic Algorithm in education.

Regional economic dispatch analysis and industry chain optimization based on proximal strategy optimization algorithm

Abstract In this paper, the traditional proximal strategy optimization algorithm is improved based on traction technology to address the poor efficiency of the conventional proximal strategy optimization algorithm in analyzing regional economic scheduling and industrial chain optimization. The focus is on finding the optimal solution by applying this algorithm in the spatio-temporal coupling characteristics of new infrastructure and regional economic dispatch, the simulation of changes in regional industrial structure and the allocation of industrial chain. The experimental analysis not only verifies the effectiveness of the improved algorithm, but also shows that the level of new infrastructure development shows a specific positive correlation with the degree of development of regional economic dispatch, with a growth rate of 29.26% in 2021. Under the influence of different industrial scenarios and policies, there are significant differences in the path and speed of industrial structure change FPPO-QP algorithm is also above 44Mdps in V2I capacity performance under the interference of 3.5×1060Bytes. In contrast, the performance of the FPPO-LP algorithm has declined. Therefore, the improved algorithm can effectively apply to regional economic dispatch analysis and industrial chain optimization.

A new analytical algorithm for uncertain fractional differential equations in the fuzzy conformable sense

<abstract><p>This paper aims to explore and examine a fractional differential equation in the fuzzy conformable derivative sense. To achieve this goal, a novel analytical algorithm is formulated based on the Laplace-residual power series method to solve the fuzzy conformable fractional differential equations. The methodology being used to discover the fuzzy solutions depends on converting the desired equations into two fractional crisp systems expressed in $ \wp $-cut form. The main objective of our algorithm is to transform the systems into fuzzy conformable Laplace space. The transformation simplifies the system by reducing its order and turning it into an easy-to-solve algorithmic equation. The solutions of three important applications are provided in a fuzzy convergent conformable fractional series. Both the theoretical and numerical implications of the fuzzy conformable concept are explored about the consequential outcomes. The convergence analysis and theorems of the developed algorithm are also studied and analyzed in this regard. Additionally, this article showcases a selection of results through the use of both two-dimensional and three-dimensional graphs. Ultimately, the findings of this study underscore the efficacy, speed, and ease of the Laplace-residual power series algorithm in finding solutions for uncertain models that arise in various physical phenomena.</p></abstract>

Forward feature selection: empirical analysis

Feature selection is an important preprocessing step in many data science and machine learning applications. Although there exist several sophisticated feature selection algorithms, their benefits are sometimes overshadowed by their complexity and slow execution. Therefore, in many cases, a more simple algorithm may be better suited. In this paper, we demonstrate that a rudimentary forward selection algorithm can achieve optimal performance with a low time complexity. Our study is based on an extensive empirical evaluation of the forward feature selection algorithm in the context of linear regression. Concretely, we compare the forward selection algorithm against the gold standard exhaustive search algorithm based on several datasets. The results show that the forward selection algorithm achieves high performance with relatively fast execution. Given the simplicity, accuracy, and speed of the forward feature selection algorithm, we recommend it as a primary feature selection method for most regression applications. Our results are particularly pertinent in the case of big data and real-time analysis.

Optical phased array output beam calibration method based on Adam algorithm

<sec>Optical phased array (OPA) technology, as a pivotal component of laser detection and ranging (LiDAR) systems, plays a crucial role in augmenting the application efficiency in fields such as autonomous driving, precision measurement, and remote sensing detection. With the escalating demands for high-resolution imaging, the array size of OPAs is continuously expanding, imposing higher requirements on the calibration precision and efficiency of the output beam. Existing calibration algorithms, such as the simultaneous perturbation stochastic gradient descent (SPGD) and the Gerchberg-Saxton (GS) algorithm, often face challenges of prolonging calibration times and insufficient precision when dealing with large-scale OPA systems.</sec><sec>In order to address this problem, our study introduces the Adam optimization algorithm, renowned for its adaptive learning rate feature, into the calibration process of OPA output beams. Through simulation modeling and experimental validation, this work comprehensively examines the differences in performance between the Adam algorithm and conventional SPGD and GS algorithms in beam calibration, especially under various OPA array configurations. For a 16×16 OPA array, the application of the Adam algorithm significantly enhances the peak side lobe ratio (PSLR) to over 15.98 dB, while notably reducing the number of iterations to less than 600, thereby shortening the calibration cycle and improving calibration precision effectively.</sec><sec>Furthermore, this work provides an in-depth analysis of parameter selection, convergence speed, and stability of the Adam algorithm in OPA calibration, offering detailed guidance for achieving more efficient and high-quality beam calibration. Through comparative analysis, this work not only demonstrates the substantial advantages of the Adam algorithm in enhancing OPA calibration efficiency, reducing calibration duration, and optimizing output beam quality but also emphasizes its critical role in advancing OPA technology.</sec><sec>The main contribution of this work lies in providing an innovative algorithmic approach for achieving efficient calibration of OPA output beams, which has important theoretical and practical significance for advancing the LiDAR technology, particularly in the field of high-precision beam control. Moreover, by applying optimized algorithms, this study not only improves the performance of OPA technology within existing domains but also paves new ways for its application in emerging fields such as optical communication, optical networking, and high-resolution imaging.</sec>

Fuzzy C Means Clustering Coupled with Firefly Optimization Algorithm for the Segmentation of Neurodisorder Magnetic Resonance Images

Medical image segmentation is a critical task in medical image analysis, and clustering algorithms can be utilized to achieve this goal. This research work focuses on the segmentation of neuro disorder magnetic resonance images using Otsu, K-means, and FCM coupled with the firefly optimization algorithm. Otsu is a classical thresholding algorithm that relies on a single threshold value to segment the images. K-Means is one of the simplest and most widely used clustering algorithms. It aims to partition data into K clusters, where each data point belongs to the cluster with the nearest mean. Fuzzy C-Means is an extension of K-Means, allowing data points to belong to multiple clusters with varying degrees of membership. In medical image segmentation, FCM was used to classify pixels or voxels into different tissue classes with soft boundaries, accounting for partial volume effects. Firefly Optimization helps in improving the convergence speed of the FCM algorithm. Firefly optimization is good at exploring the solution space and finding global optima. The combination of FCM and Firefly Optimization leads to more accurate clustering results. The performance evaluation was done by Renyi entropy and Shannon entropy, FCM coupled with the firefly optimization was found to exhibit superior results when compared with the Otsu and K-means clustering algorithms.

A numerical simulation research on fish adaption behavior based on deep reinforcement learning and fluid–structure coupling: Implementation of the “perceive-feedback-memory” control system

The autonomous swimming of fish in a complex flow environment is a nonlinear and intricate system, which is the focus and challenge in various fields. This study proposed a novel simulation framework for artificial intelligence fish. It employed a high-precision immersed boundary-lattice Boltzmann coupling scheme to simulate the interactions between fish and flow in real time, and utilized the soft actor-critic (SAC) deep reinforcement learning algorithm for fish brain decision-making module, which was further divided into a vision-based directional navigation and a lateral line-based flow perception modules, each matched with its corresponding macro-action space. The flow features were extracted using a deep neural network based on a multi-classification algorithm from the data perceived by the lateral line and were linked to the fish actions. The predation swimming and the various Kármán gait swimming were explored in terms of training, simulation, and generalization. Numerical results demonstrated significant advantages in the convergence speed and training efficiency of the SAC algorithm. Owing to the closed-loop “perceive-feedback-memory” mode, intelligent fish can respond in real-time to changes in flow fields based on reward-driven requirements and experience, and the accumulated experience can be directly utilized in other flow fields, and its adaptability, model training efficiency, and generalization were substantially improved.

Control of a neural network for the operation of small aircraft for agricultural purposes

We will discuss the impact of various parameters, such as network size and complexity, on accuracy, speed, and efficiency, as well as ways to optimize them. We also analyze the use of deep learning algorithms to detect and classify objects and adjust the navigation of small agricultural aircraft depending on the environment. We also highlight the importance of data preprocessing and how it can be used to improve the accuracy and speed of deep learning algorithms. Finally, we will discuss how deep learning can be used to automate certain tasks, such as detecting and repairing leaks or anomalies, as well as the use of aircraft from other fields, in the field of agriculture.

Application of Intelligent Algorithms for Optimal Allocation of Accounting Education Resources in Higher Education Environments

Abstract Inadequate allocation of accounting education resources and uneven spatial distribution are the key issues affecting the high-quality development of accounting education, and the arrival of the era of intelligent technology provides ideas for the rational allocation of higher education resources and transformation and upgrading. This paper constructs a multi-objective planning model for accounting education resource allocation and explores the solution method based on the fuzzy particle swarm algorithm. Population initialization is achieved by combining input-output fuzzification processing and particle swarm optimization search in fuzzy control. Subsequently, the fitness of every particle is assessed. Next, according to the formula of PSO algorithm update iteration, the position and speed are updated, the optimal solution of itself and the optimal solution of the population are updated, and the optimization search is continuously iterated. After the model is put into use the institutions in the western region with low and lower levels of efficiency in the allocation of accounting education resources decreased by 7%, and all the regions with low levels of efficiency have decreased. The number of configuration rounds of this paper’s method is smaller than the number of rounds of configuration conflicts of the PCA method and SVM method, which indicates that this paper’s method has good performance and optimizes the allocation of accounting education resources, making accounting education develop in a balanced way.

PMSM Disturbance Resistant Adaptive Fast Super-Twisting Algorithm Speed Control Method

PMSM Disturbance Resistant Adaptive Fast Super-Twisting Algorithm Speed Control Method

Chilled sampling for uncertainty quantification: a motivation from a meteorological inverse problem *

Atmospheric motion vectors (AMVs) extracted from satellite imagery are the only wind observations with good global coverage. They are important features for feeding numerical weather prediction (NWP) models. Several Bayesian models have been proposed to estimate AMVs. Although critical for correct assimilation into NWP models, very few methods provide a thorough characterization of the estimation errors. The difficulty of estimating errors stems from the specificity of the posterior distribution, which is both very high dimensional, and highly ill-conditioned due to a singular likelihood, which becomes critical in particular in the case of missing data (unobserved pixels). Motivated by this difficult inverse problem, this work studies the evaluation of the (expected) estimation errors using gradient-based Markov chain Monte Carlo (MCMC) algorithms. The main contribution is to propose a general strategy, called here ‘chilling’, which amounts to sampling a local approximation of the posterior distribution in the neighborhood of a point estimate. From a theoretical point of view, we show that under regularity assumptions, the family of chilled posterior distributions converges in distribution as temperature decreases to an optimal Gaussian approximation at a point estimate given by the maximum a posteriori, also known as the Laplace approximation. Chilled sampling therefore provides access to this approximation generally out of reach in such high-dimensional nonlinear contexts. From an empirical perspective, we evaluate the proposed approach based on some quantitative Bayesian criteria. Our numerical simulations are performed on synthetic and real meteorological data. They reveal that not only the proposed chilling exhibits a significant gain in terms of accuracy of the AMV point estimates and of their associated expected error estimates, but also a substantial acceleration in the convergence speed of the MCMC algorithms.

Optimization of Leaky-ESN parameters based on dragonfly and whale algorithm

In this research, we present a spiral flying mode-based dragonfly algorithm (DA-WOA). Since the traditional Dragonfly algorithm (DA) has a flight, the final results obtained are random in nature and have poor global search capability. Changing to a spiral flight pattern not only enlarges the search area but also makes it easy to find the optimal solution. The Swish function is also used to optimize the various flight coefficients of the DA to avoid the gradient descent approaching 0. Adding a fitness coefficient to the final position updating equation can better control the distance between dragonflies and natural enemies and thus find the optimal solution. The convergence speed of the DA-WOA algorithm and the superiority of the global search are verified by 13 test functions. In this paper, DA-WOA is applied to Leaky-ESN to optimize three parameters of the Leaky-ESN system, namely, the leakage rate (α), the spectral radius of the connectivity weight matrix (ρ), and the input scale factor (s in ). The experimental results demonstrate that DA-WOA has very strong prediction ability as well as very quick convergence speed for the predicted Sin and MG time series.