Convergence Rates For Problems Research Articles

AQMDRL: Automatic Quality of Service Architecture Based on Multistep Deep Reinforcement Learning in Software-Defined Networking

Software-defined networking (SDN) has become one of the critical technologies for data center networks, as it can improve network performance from a global perspective using artificial intelligence algorithms. Due to the strong decision-making and generalization ability, deep reinforcement learning (DRL) has been used in SDN intelligent routing and scheduling mechanisms. However, traditional deep reinforcement learning algorithms present the problems of slow convergence rate and instability, resulting in poor network quality of service (QoS) for an extended period before convergence. Aiming at the above problems, we propose an automatic QoS architecture based on multistep DRL (AQMDRL) to optimize the QoS performance of SDN. AQMDRL uses a multistep approach to solve the overestimation and underestimation problems of the deep deterministic policy gradient (DDPG) algorithm. The multistep approach uses the maximum value of the n-step action currently estimated by the neural network instead of the one-step Q-value function, as it reduces the possibility of positive error generated by the Q-value function and can effectively improve convergence stability. In addition, we adapt a prioritized experience sampling based on SumTree binary trees to improve the convergence rate of the multistep DDPG algorithm. Our experiments show that the AQMDRL we proposed significantly improves the convergence performance and effectively reduces the network transmission delay of SDN over existing DRL algorithms.

Coarse-to-fine feature representation based on deformable partition attention for melanoma identification

In the histopathological melanoma image diagnosis system, manual identification of super-scale slides with dense cells is tedious, time-consuming, and subjective. To deal with this problem, we propose an automatic identification network based on the deformable partition attention to identify lots of dense slides as an assistant. A coarse-to-fine strategy is adopted in feature representation and qualitative identification to improve the identification accuracy of melanomas and nevi. First of all, because it is difficult to extract features in the lesion area with blurred boundaries and uneven distribution, we develop a deformable partition attention module, which integrates the advantage of the attention mechanism and deformable convolution. The module overcomes the limitation of rectangular convolution and gradually refines the channel and spatial features, which enriches feature representation by combining global and local features. Secondly, to address the problem of difficult convergence and poor recognition rate caused by the excessive non-aligned distance between benign-malignant and benign subcategories, we propose a progressive architecture via a coarse sub-network closely followed by a fine sub-network. Moreover, to further increase the inter-class differences and reduce the intra-class disparities, we propose a joint loss function to mine hard samples, which effectively improves the identification performance. Experimental results on the clinical dataset show that the proposed algorithm has higher sensitivity and specificity and outperforms state-of-the-art deep neural networks.

H∞-Based Minimal Energy Adaptive Control With Preset Convergence Rate.

This work studies the H∞ -based minimal energy control with a preset convergence rate (PCR) problem for a class of disturbed linear time-invariant continuous-time systems with matched external disturbance. This problem aims to design an optimal controller so that the energy of the control input satisfies a predetermined requirement. Moreover, the closed-loop system asymptotic stability with PCR is ensured simultaneously. To deal with this problem, a modified game algebraic Riccati equation (MGARE) is proposed, which is different from the game algebraic Riccati equation in the traditional H∞ control problem due to the state cost being lost. Therefore, a unique positive-definite solution of the MGARE is theoretically analyzed with its existing conditions. In addition, based on this formulation, a novel approach is proposed to solve the actuator magnitude saturation problem with the system dynamics being exactly known. To relax the requirement of the knowledge of system dynamics, a model-free policy iteration approach is proposed to compute the solution of this problem. Finally, the effectiveness of the proposed approaches is verified through two simulation examples.

H-Matrix-Based Direct Solver of JMCFIE for the Analysis of Scattering From Penetrable Objects

Method of moments (MoMs) solution of the electric and magnetic current combined-field integral equation (JMCFIE) for penetrable objects generally resorts to iterative solvers, which suffer from the problems of convergence rate and redundant computation for multiple right-hand-sides (RHSs). In this communication, an efficient direct solver based on the hierarchical ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mathcal {H}$ </tex-math></inline-formula> -) matrix technique is presented to solve these problems. <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mathcal {H}$ </tex-math></inline-formula> -matrix formatted algorithms provide an efficient way to factorize the impedance matrix with low computational costs, which make the direct solution feasible. Examples of electromagnetic scattering from a dielectric object and optical scattering from a plasmonic nanostructure are shown to demonstrate the efficiency and accuracy of the proposed method.

Isogeometric analysis based on modified Loop subdivision surface with improved convergence rates

This paper introduces a modified version of Loop subdivision, called modified Loop subdivision surface (MLSS), to improve the convergence rates in isogeometric analysis. Motivated by the control net of B-splines with non-repetitive end knots, the subdivision surface is parameterized on the non-boundary elements, where no special subdivision rules and evaluation needed for the boundaries. This idea together with the Nitsche’s method which is adopted for weakly imposing boundary conditions is used to solve the non-optimal convergence rate problem for extended Loop subdivision even without extraordinary points (EPs). For the EPs, we define a new rule with a parameter λ (0<λ<1.0), which is exactly the second maximum eigenvalue of the subdivision matrix. The MLSS is defined by combining the two ideas, which is global C2-continuous except G1 and curvature bounded around the EPs. The MLSS limit surface has comparable shape quality as Loop subdivision surface. Besides that, the numerical experiments show that MLSS can achieve the optimal convergence rate in the Poisson problem both in the L2 and H1 norm.

New Solutions of Nonlinear Equations Constrained by Bounded Variables under Edge Computation

Aiming at the problem of low global convergence and local convergence rate of trust region interior points of bounded variable constrained nonlinear equations, a trust region interior point algorithm for bounded variable constrained nonlinear equations under edge calculation is designed. By constructing the basic function form of nonlinear equations constrained by bounded variables, the boundary of nonlinear equations is determined by Gauss Newton iterative process to ensure the global convergence of changes; solve the original objective function, analyze the trust region subproblem of the unconstrained optimization problem, and generate an acceptable region. The region is generated through two-dimensional example interpretation. The interior points in the acceptable region are determined by the primal dual interior point method, and the interior points in the acceptable region are optimized. With the help of edge calculation, the trust region interior point programming model of bounded variable constrained nonlinear equations is designed to realize the algorithm design. The experimental results show that the designed algorithm can improve the trust region interior point global convergence and local convergence rate of nonlinear equations with bounded variable constraints.

Fast and Accurate Prediction of Light Scattering from Plasmonic Nanoarrays in Multiple Directions.

The method of moments (MoM) is an efficient electromagnetic numerical method for the accurate prediction of light scattering from plasmonic nanostructures. In practice, the light-scattering properties in different incident directions are often concerning. However, traditional MoM generally resorts to the iterative method, which suffers from the problems of convergence rate and redundant computations for multiple incident excitations. Nanoarray structures will further aggravate these problems due to a large number of unknowns. In this article, an efficient numerical method based on MoM and a hierarchical matrix (H-matrix) algorithm is proposed to solve these problems. Numerical experiments demonstrate the efficiency and accuracy of the proposed method for the prediction of light scattering from plasmonic nanoarrays in multiple directions.

Improved slime mould algorithm with elitist strategy and its application to structural optimization with natural frequency constraints

An improved slime mould algorithm (ISMA) is proposed and applied to the size optimization of truss structures with natural frequency constraints. The slime mould algorithm is a recently proposed metaheuristic inspired by the morphological changes of the acellular slime mould Physarum polycephalum while foraging. This algorithm has been successfully applied to some real-world optimization problems in science and industry. However, it is found that the classical SMA suffers from slow convergence and often converges prematurely to non-optimal solutions, especially for large-scale optimization problems. Compared to the classical SMA, two main improvements are introduced in the proposed ISMA: (1) In the replacement phase of the ISMA, an elitist strategy is adopted to replace the generational replacement strategy of the classical SMA. The elitist strategy aims to increase the convergence rate of the ISMA. (2) A slight modification is made to the exploration phase of the classical SMA to ensure a vast exploration of the search space. The suggested modification enables the ISMA to overcome the shortcoming of the premature convergence of the classical SMA. Finally, the efficiency and robustness of the ISMA are demonstrated through three large-scale benchmark dome trusses with natural frequency constraints. To our knowledge, this is the first time to apply SMA to structural optimization. It is shown that the proposed ISMA overcomes the problems of premature convergence and slow convergence rate of the classical SMA. Numerical results show that the ISMA outperforms the classical SMA and shows superior or comparable performance to other reported methods.

Semi-Airborne electromagnetic 2.5D inversion based on a PSO–LCI strategy

The processing of semi-airborne electromagnetic data (SAEM) mainly adopts1D or quasi-2D inversion, which is not conducive to fine detection. By targeting 2.5D inversion problem of SOTEM, a hybrid algorithm based on particle swarm optimization (PSO) and lateral constraint inversion (LCI) is optimized. PSO can remove dependency on the starting model and improve the accuracy of interpretation. LCI may overcome the disadvantages of PSO, such as low convergence rate and over-stochastic problem. The cosine probability function varying with the number of iterations is used to merge the two algorithms to arrive at a compromise between precision and speed. Compared with the particle swarm optimization algorithm, the convergence speed of the hybrid algorithm is significantly accelerated in the late stage and successfully suppresses the over-stochastic problem. Lastly, the new algorithm is applied to the inversion of the data measured by the coal mine target.

A soft computing based multi-objective optimization approach for automatic prediction of software cost models

This paper tries to extend the idea of single-objective differential evolution (DE) algorithm to a multi-objective algorithm. Most of the existing algorithms face the problem of diversity loss and convergence rate. In this paper, we propose a novel multi-objective DE algorithm to deal with this problem. In the validation process, the proposed method is validated in two steps. Firstly, the new homeostasis factor-based mutation operator incorporates multi-objective differential evolution algorithms (MODE). In this method, we use the Pareto optimality principle. We incorporate a new adaptive-based mutation operator (MODE) to create more diversity and enhance convergence rate among candidate solutions which provide better solutions to help the evolution. The effectiveness of the proposed method is evaluated on eight benchmarks of bi-objective and tri-objective test functions. Our proposed method performed well compared to the latest variants of multi-objective evolutionary algorithms (MOEAs). Secondly, the proposed method is used for an application-based test by applying it for software cost estimation. This method also incorporates multi-objective parameters, i.e., two objectives-based software cost estimation and three objectives-based software cost estimation. The proposed approach achieves better results in most software projects in terms of reducing effort and minimum error.

Improved Salp Swarm Algorithm with Simulated Annealing for Solving Engineering Optimization Problems

Swarm-based algorithm can successfully avoid the local optimal constraints, thus achieving a smooth balance between exploration and exploitation. Salp swarm algorithm (SSA), as a swarm-based algorithm on account of the predation behavior of the salp, can solve complex daily life optimization problems in nature. SSA also has the problems of local stagnation and slow convergence rate. This paper introduces an improved salp swarm algorithm, which improve the SSA by using the chaotic sequence initialization strategy and symmetric adaptive population division. Moreover, a simulated annealing mechanism based on symmetric perturbation is introduced to enhance the local jumping ability of the algorithm. The improved algorithm is referred to SASSA. The CEC standard benchmark functions are used to evaluate the efficiency of the SASSA and the results demonstrate that the SASSA has better global search capability. SASSA is also applied to solve engineering optimization problems. The experimental results demonstrate that the exploratory and exploitative proclivities of the proposed algorithm and its convergence patterns are vividly improved.

Predicting LncRNA-Disease Association Based on Generative Adversarial Network.

Increasing research reveals that long non-coding RNAs (lncRNAs) play an important role in various biological processes of human diseases. Nonetheless, only a handful of lncRNA-disease associations have been experimentally verified. The study of lncRNA-disease association prediction based on the computational model has provided a preliminary basis for biological experiments to a great degree so as to cut down the huge cost of wet lab experiments. This study aims to learn the real distribution of lncRNA-disease association from a limited number of known lncRNA-disease association data. This paper proposes a new lncRNA-disease association prediction model called LDA-GAN based on a Generative Adversarial Network (GAN). Aiming at the problems of slow convergence rate, training instabilities, and unavailability of discrete data in traditional GAN, LDA-GAN utilizes the Gumbel-softmax technology to construct a differentiable process for simulating discrete sampling. Meanwhile, the generator and the discriminator of LDA-GAN are integrated to establish the overall optimization goal based on the pairwise loss function. Experiments on standard datasets demonstrate that LDA-GAN achieves not only high stability and high efficiency in the process of confrontation learning but also gives full play to the semisupervised learning advantage of generative adversarial learning framework for unlabeled data, which further improves the prediction accuracy of lncRNA-disease association. Besides, case studies show that LDA-GAN can accurately generate potential diseases for several lncRNAs. We introduce a generative adversarial model to identify lncRNA-disease associations.

Cooperative control with designated convergence rate for high-order integrators under heterogeneous couplings

In this paper, we investigate the cooperative control problem of high-order integrators under heterogeneous couplings. A new class of distributed control algorithms are developed for the designated convergence rate (DCR) problem of high-order integrators, which could explicitly show the convergence margin of the closed-loop system, and has better robustness than conventional consensus algorithms. We first propose state consensus control algorithms for high-order integrators, where necessary and sufficient convergence conditions are proposed by theoretical analysis. Then we extend the results to the case of output leaderless consensus of heterogeneous high-order integrators with heterogeneous couplings. Finally, simulation examples are given to validate the effectiveness of the proposed algorithms.

Cuckoo Search-based SVM (CS-SVM) Model for Real-Time Indoor Position Estimation in IoT Networks

Internet of Things (IoT), an emerging technology, is becoming an essential part of today’s world. Machine learning (ML) algorithms play an important role in various applications of IoT. For decades, the location information has been extremely useful for humans to navigate both in outdoor and indoor environments. Wi-Fi access point-based indoor positioning systems get more popularity, as it avoids extra calibration expenses. The fingerprinting technique is preferred in an indoor environment as it does not require a signal’s Line of Sight (LoS). It consists of two phases: offline and online phase. In the offline phase, the Wi-Fi RSSI radio map of the site is stored in a database, and in the online phase, the object is localized using the offline database. To avoid the radio map construction which is expensive in terms of labor, time, and cost, machine learning techniques may be used. In this research work, we proposed a hybrid technique using Cuckoo Search-based Support Vector Machine (CS-SVM) for real-time position estimation. Cuckoo search is a nature-inspired optimization algorithm, which solves the problem of slow convergence rate and local minima of other similar algorithms. Wi-Fi RSSI fingerprint dataset of UCI repository having seven classes is used for simulation purposes. The dataset is preprocessed by min-max normalization to increase accuracy and reduce computational speed. The proposed model is simulated using MATLAB and evaluated in terms of accuracy, precision, and recall with K-nearest neighbor (KNN) and support vector machine (SVM). Moreover, the simulation results show that the proposed model achieves high accuracy of 99.87%.

EFFICIENT MULTIPHYSICS ITERATIONS IN MPACT WITH PARTIALLY CONVERGENT CMFD

Partial convergence of CMFD can help to stabilize multiphysics iteration schemes. In this paper, an efficient multiphysics iteration scheme with near-optimal partially convergent CMFD implemented in MPACT is presented. In the new scheme, the feedback intensity of the problem is automatically estimated, and the relative convergence of CMFD solver is adjusted accordingly. Numerical results show that MPACT with near-optimal partially convergent CMFD can have almost the same convergence rate in problems with feedback as those without feedback. For the problems tested here the run time may be reduced by more than 20% and up to 49% compared with that of current MPACT.

One-dimension hierarchical local receptive fields based extreme learning machine for radar target HRRP recognition

Radar automatic target recognition (RATR) aims at extracting meaningful target features from the electromagnetic echo signal and utilizing the features to automatically recognize the target types. The high-resolution range profile (HRRP) plays an important role in RATR field, HRRP is the amplitude of the echo summation for target scattering centers in each range cell of wideband radar. Using deep neural networks for HRRP radar target recognition encounters the problem of storage overhead and slow convergence rate, to resolve those issues, we propose a one-dimension local receptive fields based extreme learning auto-encoder (1D ELM-LRF-AE) network for HRRP local structures and meaningful representations learning. ELM-LRF-AE consists of an input layer, a random convolution layer, a pooling layer, several local connected layers and an output layer, it reconstructs the input with a greedy strategy that the input feature vectors are divided into several subgroups and the i-th pooling feature vector is used to reconstruct the i-th grouping input feature vector. Then we use the learned pooling feature vectors to replace the random pooling feature vectors as the learned representations. We also stack several 1D ELM-LRF-AEs to build 1D hierarchical local receptive fields based extreme learning machine (1D H-ELM-LRF) for high level HRRP abstract representations learning and recognition. Experimental results on simulated HRRP data set demonstrate the superior high recognition performance and high computational efficiency of our algorithm.

The On-Chip D-LMS Filter Design Method of Wireless Sensor Node Based on FPGA

In the real-time position technology of underground shallow source, the signal denoising performance of wireless sensor nodes directly determines the location speed and accuracy of underground burst point. Because of the complexity and randomness of the underground medium and the fact that underground explosion is a nonstationary transient process, the problems of low convergence rate and poor steady-state performance of the filter exist when the existing LMS algorithm is used for signal denoising. In light of the above concerns, this paper comes up with a signal denoising algorithm and hardware implementation method based on D-LMS (delay-LMS). Firstly, according to the autocorrelation function characteristic of random signal, using the principle that the autocorrelation function time delay characteristic of narrowband signal such as explosion vibration signal is better than that of wideband random signal such as ground noise, the D-LMS filter algorithm is constructed by introducing the time delay parameter. Secondly, the selection method of key parameters in D-LMS hardware implementation is analyzed. Thirdly, the corresponding hardware circuit is designed by FPGA, and the simulation is carried out. Numerical simulation and experimental verification show that compared with the existing LMS improved algorithm, the D-LMS algorithm proposed in this paper has higher denoising stability and better denoising effect. Compared with the signal postprocessing method based on the host computer, the signal denoising speed of this method is significantly improved. This method will provide a powerful theoretical method to solve the problem of high-precision and fast source positioning and provide technical support for the development of high-speed and real-time source positioning instruments.

Short-term traffic flow prediction based on improved wavelet neural network

Due to the characteristics of time-varying traffic and nonlinearity, the short-term traffic flow data are difficult to predict accurately. The purpose of this paper is to improve the short-term traffic flow prediction accuracy through the proposed improved wavelet neural network prediction model and provide basic data and decision support for the intelligent traffic management system. In view of the extremely strong nonlinear processing power, self-organization, self-adaptation and learning ability of wavelet neural network (WNN), this paper uses it as the basic prediction model and uses the particle swarm optimization algorithm for the slow convergence rate and local optimal problem of WNN prediction algorithm. With the advantages of fast convergence, high robustness and strong global search ability, an improved particle swarm optimization algorithm is proposed to optimize the wavelet neural network prediction model. The improved wavelet neural network is used to predict short-term traffic flow. The experimental results show that the proposed algorithm is more efficient than the WNN and PSO–WNN algorithms alone. The prediction results are more stable and more accurate. Compared with the traditional wavelet neural network, the error is reduced by 14.994%.

Research on Dynamic Path Planning of Wheeled Robot Based on Deep Reinforcement Learning on the Slope Ground

The existing dynamic path planning algorithm cannot properly solve the problem of the path planning of wheeled robot on the slope ground with dynamic moving obstacles. To solve the problem of slow convergence rate in the training phase of DDQN, the dynamic path planning algorithm based on Tree-Double Deep Q Network (TDDQN) is proposed. The algorithm discards detected incomplete and over-detected paths by optimizing the tree structure, and combines the DDQN method with the tree structure method. Firstly, DDQN algorithm is used to select the best action in the current state after performing fewer actions, so as to obtain the candidate path that meets the conditions. And then, according to the obtained state, the above process is repeatedly executed to form multiple paths of the tree structure. Finally, the non-maximum suppression method is used to select the best path from the plurality of eligible candidate paths. ROS simulation and experiment verify that the wheeled robot can reach the target effectively on the slope ground with moving obstacles. The results show that compared with DDQN algorithm, TDDQN has the advantages of fast convergence and low loss function.

Evaluation of road condition based on BA-BP algorithm

BP (Back Propagation) neural network has been widely applied for classification tasks including road condition evaluation, however, BP model has the problem of lower accuracy and slow convergence rate. A novel road condition evaluation method based on BA-BP (Bat-Back Propagation) algorithm is proposed for the unstructured small road condition evaluation, which filled the vacancy of specific small road scenes. Firstly, five kinds of road condition features including roughness, curvature, obstacle width to height ratio, obstacle effective area ratio, obstacle coefficient are defined and extracted. Then obstacles from region of interest (ROI) in front of the vehicle are analyzed. Finally, Bat algorithm is used to optimize the searching of initial network weights and thresholds, which obtained a higher accuracy of 95.15% and efficient training process. Comparison experiments showed that the proposed approach improved the accuracy with 5.31%, 3.32%, 3.17% than the BP, GA-BP and FA-BP model, respectively. As for the processing time of collected road data, BA-BP network consumed less time of 2 s and 3.9 s compared with GA-BP and FA-BP. Proposed method also outperformed than most of the state-of-the-art approaches with higher accuracy and simpler hardware environments, which proved its potential of being popularized in large scale real-time systems.