Feasibility Of Algorithm Research Articles

Where Reinforcement Learning Meets Process Control: Review and Guidelines

This paper presents a literature review of reinforcement learning (RL) and its applications to process control and optimization. These applications were evaluated from a new perspective on simulation-based offline training and process demonstrations, policy deployment with transfer learning (TL) and the challenges of integrating it by proposing a feasible approach to online process control. The study elucidates how learning from demonstrations can be accomplished through imitation learning (IL) and reinforcement learning, and presents a hyperparameter-optimization framework to obtain a feasible algorithm and deep neural network (DNN). The study details a batch process control experiment using the deep-deterministic-policy-gradient (DDPG) algorithm modified with adversarial imitation learning.

Early Diagnosis of Chemotherapy-Linked Cardiotoxicity in Breast Cancer Patients Using Conventional Biomarker Panel: A Prospective Study Protocol

The prognosis of cancer treatment depends on, among other aspects, the cardiotoxicity of chemotherapy. This research aims to create a feasible algorithm for the early diagnosis of antitumor therapy cardiotoxicity in breast cancer patients. The paper represents a protocol for a prospective cohort study with N 120 eligible participants admitted for treatment with anthracyclines and/or trastuzumab. These patients will be allocated into four risk groups regarding potential cardiotoxic complications. Patients will be examined five times every three months for six biomarkers,: cardiac troponin I (cTnI), brain natriuretic peptide (BNP), C-reactive protein (CRP), myeloperoxidase (MPO), galectin-3 (Gal-3), and D-dimer, simultaneously with echocardiographic methods, including speckle tracking. The adjusted relative risk (aOR) of interrupting an entire course of chemotherapy due to cardiotoxic events will be assessed using multiple analyses of proportional Cox risks. The Cox model will also assess associations between baseline biomarker values and time to cardiotoxic events. Moreover, partly conditional survival models will be applied to determine associations between repeated assessments of changes in biomarkers from baseline and time to cancer therapy-related cardiac dysfunction. All models will be adjusted for cancer therapy regimen, baseline LVEF, groups at risk, baseline biomarker values, and age. The decision-tree and principal component analysis (PCA) methods will also be applied. Thus, feasible patterns will be detected.

A class of new search directions for full-NT step feasible interior point method in semidefinite optimization

In this paper, based on Darvay et al.’s strategy for linear optimization (LO) (Z. Darvay and P.R. Takács, Optim. Lett. 12 (2018) 1099–1116.), we extend Kheirfam et al.’s feasible primal-dual path-following interior point algorithm for LO (B. Kheirfam and A. Nasrollahi, Asian-Eur. J. Math. 1 (2020) 2050014.) to semidefinite optimization (SDO) problems in order to define a class of new search directions. The algorithm uses only full Nesterov-Todd (NT) step at each iteration to find an ε-approximated solution to SDO. Polynomial complexity of the proposed algorithm is established which is as good as the LO analogue. Finally, we present some numerical results to prove the efficiency of the proposed algorithm.

An Effective Approach for Smart Parking Management

Drivers and motorists get annoyed when it takes a long time to find a vacant space in a parking lot. Looking for parking has become a headache as the number of vehicles in urban cities and the cost of land concurrently increase. There is an urgent need for innovation in smart parking systems. Currently, investors and contractors pay laborers to operate and maintain smart parking systems. Staff duties may include opening and closing gates, giving directions to drivers and motorists, and managing payments associated with the lot. This article proposes a feasible, dependable, and smart algorithm for managing a parking system. This algorithm utilizes image processing techniques to provide real-time data. No labor is required to operate and handle the system. The system itself automatically handles all operations except maintenance. Furthermore, this algorithm is more cost-effective than other similar systems and equally effective. Numerous simulation scenarios were carried out on MATLAB to verify its developed approach. A comparison evaluation juxtaposes the proposed approach with other solutions in the literature. This evaluation clearly indicates that the presented method outperforms other solutions in terms of technologies being used, devices being utilized, and cost.

ECG Classification Based on Wasserstein Scalar Curvature

Electrocardiograms (ECG) analysis is one of the most important ways to diagnose heart disease. This paper proposes an efficient ECG classification method based on Wasserstein scalar curvature to comprehend the connection between heart disease and the mathematical characteristics of ECG. The newly proposed method converts an ECG into a point cloud on the family of Gaussian distribution, where the pathological characteristics of ECG will be extracted by the Wasserstein geometric structure of the statistical manifold. Technically, this paper defines the histogram dispersion of Wasserstein scalar curvature, which can accurately describe the divergence between different heart diseases. By combining medical experience with mathematical ideas from geometry and data science, this paper provides a feasible algorithm for the new method, and the theoretical analysis of the algorithm is carried out. Digital experiments on the classical database with large samples show the new algorithm’s accuracy and efficiency when dealing with the classification of heart disease.

Intelligent Virtual Resource Allocation of QoS-Guaranteed Slices in B5G-Enabled VANETs for Intelligent Transportation Systems

5G communication technologies and networks help researchers and engineers look into intelligent transportation systems (ITS) with a new eye, including vehicular ad hoc networks (VANET) application. Network function virtualization (NFV) and network slicing (NS) are accepted as two most promising technologies towards the agile and elastic network architecture of 5G and beyond 5G (B5G). However, previous researchers studied NFV and NS separately. In addition, learning technologies, such as reinforcement leaning (RL), graph-based learning, emerge so as to enhance the network intelligence and resource allocation in recent years. Inspired from these, we jointly explore intelligent resource allocation issue within B5G-enabled VANETs. At first, the novel virtual resource allocation framework supporting NFV and NS for providing quality of service (QoS)-guaranteed slices is constructed. Then, we formulate the virtual resource allocation of slices as the optimization problem, having the goals of providing guaranteed QoS performance and maximizing the net profit. Considering the non convex attributes of the formulated optimization problem, we propose one intelligent and feasible algorithm instead, including the details of the proposed intelligent algorithm. We record the results in order to validate the feasibility and highlights of our proposed algorithm. For example, our intelligent algorithm has the slice acceptance advantage of 5%, comparing with the best existing work.

Fault-Tolerant Operation of Bidirectional ZSI-Fed Induction Motor Drive for Vehicular Applications

This paper presents an efficient and fast fault-tolerant control scheme for a bidirectional Z-source inverter (BiZSI)-fed induction-motor drive system for vehicular applications. The proposed strategy aims for the fault detection, localization and diagnosis of the proposed system during switch failures in the inverter module. Generally, power–semiconductor switch failures in inverter modules occur due to open- and short-circuit faults. An efficient modulation scheme is proposed and design specifications are thoroughly derived to obtain high voltage gains across the BiZSI network. A suitably fast detection and diagnosis scheme to isolate the faulty leg and resume the normal operation is discussed in this paper. The control scheme is provided such that the faulty leg is isolated and the motor phase is fed from a redundant leg to resume the operation. A feasible localization algorithm based on experimentally derived values and switching vectors is implemented. In addition, a fast fault diagnosis method based on current estimation and motor speed variation is designed and implemented. Moreover, the most important advantages of the proposed strategy include lower hardware requirements and less harmonic distortion in the output currents. Finally, the simulation and experimental results are presented to validate the feasibility of the theoretical analysis. An extensive performance evaluation of the proposed system with fault ride-through capabilities is performed to prove its suitability for vehicular applications. To validate its merits, the proposed strategy is compared with similar fault-tolerant schemes currently used in the industry.

Optimal subsampling for parametric accelerated failure time models with massive survival data.

With increasing availability of massive survival data, researchers need valid statistical inferences for survival modeling whose computation is not limited by computer memories. Existing works focus on relative risk models using the online updating and divide-and-conquer strategies. The subsampling strategy has not been available due to challenges in developing the asymptotic properties of the estimator under semiparametric models with censored data. This article tackles optimal subsampling algorithms to fast approximate the maximum likelihood estimator for parametric accelerate failure time models with massive survival data. We derive the asymptotic distributions of the subsampling estimator and the optimal sampling probabilities that minimize the asymptotic mean squared error of the estimator. A feasible two-step algorithm is proposed where the optimal sampling probabilities in the second step are estimated based on a pilot sample in the first step. The asymptotic properties of the two-step estimator are established. The performance of the estimator is validated in a simulation study. A real data analysis illustrates the usefulness of the methods.

Robot Trajectory Planning Based on the Energy Management Strategy

With the increasing demand for automated production technology in national defense, industry, agriculture, and other fields, the status and role of mobile robots are becoming more and more pivotal. The intelligent technology of robots is becoming more and more demanding in terms of reliability, stability, efficiency, and adaptability, and the autonomous navigation technology of mobile robots is facing new challenges. This paper introduces the basic principle of traditional A∗ algorithm, points out the problems of this algorithm such as cumbersome calculation, large turning angle, and unsmooth derived trajectory planning, and proposes an improved A∗ algorithm. The improved algorithm introduces a two-way alternating search mechanism to improve the efficiency of the search path, and the improvement of the heuristic function solves the problem that the two-way alternating search A∗ algorithm takes more time when there are obstacles perpendicular to the path on the way to the encounter. Simulation experiments in different raster environments prove that compared with the traditional A∗ algorithm, genetic algorithm, and simulated annealing algorithm, the improved A∗ algorithm can significantly improve the search path speed, while overcoming the disadvantages of many path turning angles and large turning angles and is an efficient and feasible algorithm for raster map environments.

A learning-based synthesis approach of reward asynchronous probabilistic games against the linear temporal logic winning condition.

The traditional synthesis problem is usually solved by constructing a system that fulfills given specifications. The system is constantly interacting with the environment and is opposed to the environment. The problem can be further regarded as solving a two-player game (the system and its environment). Meanwhile, stochastic games are often used to model reactive processes. With the development of the intelligent industry, these theories are extensively used in robot patrolling, intelligent logistics, and intelligent transportation. However, it is still challenging to find a practically feasible synthesis algorithm and generate the optimal system according to the existing research. Thus, it is desirable to design an incentive mechanism to motivate the system to fulfill given specifications. This work studies the learning-based approach for strategy synthesis of reward asynchronous probabilistic games against linear temporal logic (LTL) specifications in a probabilistic environment. An asynchronous reward mechanism is proposed to motivate players to gain maximized rewards by their positions and choose actions. Based on this mechanism, the techniques of the learning theory can be applied to transform the synthesis problem into the problem of computing the expected rewards. Then, it is proven that the reinforcement learning algorithm provides the optimal strategies that maximize the expected cumulative reward of the satisfaction of an LTL specification asymptotically. Finally, our techniques are implemented, and their effectiveness is illustrated by two case studies of robot patrolling and autonomous driving.

Topology optimization for 3D concrete printing with various manufacturing constraints

The integration between 3D concrete printing (3DCP) and topology optimization (TO) enables the fabrication of structurally efficient components without expensive formwork and intensive labor. However, manufacturing constraints of 3DCP are impeding the integration between the two fields, and there has been limited research on this topic. In this paper, we address various manufacturing constraints of 3DCP within the bi-directional evolutionary structural optimization (BESO) framework. Firstly, a layer-wise sensitivity scheme is proposed to generate self-supporting designs in the user-defined print direction. Secondly, a novel continuous extrusion constraint is implemented to facilitate the continuous printing operation of the design. Thirdly, the anisotropy of the 3DCP process is simulated during optimization by employing a transverse isotropic material model. Fourthly, domain segmentation is introduced to facilitate modular construction, and each partitioned segment can be assigned with its favorable print direction. Finally, the algorithm's feasibility is demonstrated by constructing a topology optimized chair.

Linear programming with a feasible direction interior point technique for smooth optimization

We propose an adaptation of the Feasible Direction Interior Points Algorithm (FDIPA) of J. Herskovits, for solving large-scale linear programs. At each step, the solution of two linear systems with the same coefficient matrix is determined. This step involves a significant computational effort. Reducing the solution time of linear systems is, therefore, a way to improve the performance of the method. The linear systems to be solved are associated with definite positive symmetric matrices. Therefore, we use Split Preconditioned Conjugate Gradient (SPCG) method to solve them, together with an Incomplete Cholesky preconditioner using Matlab’s ICHOL function. We also propose to use the first iteration of the conjugate gradient, and to presolve before applying the algorithm, in order to reduce the computational cost. Following, we then provide mathematica proof that show that the iterations approach Karush–Kuhn–Tucker points of the problem under reasonable assumptions. Finally, numerical evidence show that the method not only works in theory but is also competitive with more advanced methods.

Interior Reconstruction from Truncated Projection Data in Cone-beam Computed Tomography.

The interior reconstruction of completely truncated projection data is a frontier research hotspot in cone-beam computed tomography (CBCT) application. It is difficult to find a method with acceptable accuracy and high efficiency to solve it. Based on the simplified algebraic reconstruction technique (S-ART) algorithm and the filtered back projection (FBP) algorithm with the new filter, an efficient and feasible interior reconstruction algorithm is proposed in this paper. The algorithm uses the S-ART algorithm to quickly recover the complete projection data and then uses the new ramp filter which can suppress the high-frequency noise in the projection data to filter the recovered complete projection data. Finally, the interior reconstructed images are obtained by back projection. The computational complexity of the proposed algorithm is close to that of the FBP algorithm for the reconstruction of the whole object, and the reconstructed image quality is acceptable, which provides an effective method for interior reconstruction in CBCT. Simulation results show the effectiveness of the method.

Aerodynamic Analysis and Training Research of an S-Shaped Arc Ball Based on Hydrodynamics.

The aerodynamic characteristics of S-shaped arc ball, such as poor combat performance and low intelligence, are studied and analyzed. In this paper, the aerodynamic analysis model of S-shaped arc soccer intelligent algorithm is established, and the dynamic tracking model of Soccer Tactics Based on fluid dynamics search algorithm is designed. The data information is collected from many aspects, such as the position of the players, the change of the arc ball movement, the trajectory of the football movement, and tactical flexibility. The results show that Benn's algorithm can start the layout mechanism effectively. It has high feasibility and accurate algorithm accuracy and can effectively improve the aerodynamic layout performance of the system. The aerodynamic analysis and training optimization method of the S-shaped arc ball based on ant colony optimization has sped up the intelligent training system of Chinese football tactics.

A wonderful triangle in compressed sensing

In order to explore some concrete metric relationship between the ‖x‖1/‖x‖∞ and ‖x‖0, we introduce a wonderful triangle whose sides are composed of ‖x‖0,‖x‖1 and ‖x‖∞ for any non-zero vector x∈Rn by delving into the iterative soft-thresholding operator in this paper. Based on the angle β of the triangle corresponding to the side whose length is ‖x‖∞-‖x‖1/‖x‖0, we demonstrate that the sparsity of signal is generally meaningful within a certain exact interval [1,50] without regard to its dimension. Finally, we construct a feasible algorithm for solving ℓ1/ℓ∞ minimization, which further illustrates that it is effective for sparse recovery.

Dynamic interception point guidance algorithm based on particle swarm optimization

The engagement of target-interceptor is an extremely complicated and nonlinear problem. Most literatures of developed guidance algorithms are hard to work in real-time missile guidance systems because of the complicated design of controllers, restriction in specific condition or excess computing loading. In this paper, the proposed guidance algorithm computes the predicted interception point of target-interceptor and applies particle swarm optimization to optimize the lateral acceleration control commands of missile where the definition of fitness function can guide the missile toward the predicted interception point when the computed fitness value is the minimum. According to the results of simulation experiments, the proposed method has the satisfied target-kill performance to the superior aircraft with high agility. The missile can greatly revise the flight route toward the computed collision course at the initial pursuit stage and the course curve of missile is flatter than the other two guidance laws. Besides, the proposed method can reduce the occurrence of big lateral acceleration control commands acting on the missile to avoid unlocking the evasive target at the terminal stage. As a result, the proposed guidance algorithm based on particle swarm optimization is very effective without using the complicated nonlinear control methods and excess storage burden of computer. It is a simple and feasible missile guidance algorithm due to the advantages of simplicity and effectiveness just like the proportional navigation guidance law but the performance of proposed guidance algorithm is better than proportional navigation guidance law and the other guidance algorithm designed by particle swarm optimization.

Prediction of Feasibility of Entrepreneurial Proposals in Student Creativity Program

Student Creativity Program is a program organized by the Directorate of Learning and Student Affairs, Directorate General of Higher Education, Research, and Technology, Ministry of Education, Culture, Research and Technology as a national level student creativity event as an effort to grow, accommodate, and realize students' creative and innovative ideas. Based on 2017-2021 data, each year an average of 63,337 proposals are received, administrative and substance evaluations involve complex assessment components and are carried out manually so that it takes a relatively long time in the calculation process. Then a special method is needed that speeds up the processing of assessment data. This research was conducted on the substance of the Entrepreneurship Sector to predict the feasibility of a proposal to get funding applying data mining with the Naive Bayes Classifier (NBC) and K-Nearest Neighbor (K-NN) algorithms with a comparison between Euclidean Distance and Manhattan Distance. From the results, it is known that NBC produces 96.49% accuracy and 0.912 Kappa. K-NN with the largest Euclidean Distance calculation in K-5, K-7 and K9 with an accuracy of 99.04% and Kappa 0.975, K-NN Manhattan Distance calculation produces the greatest accuracy of all the methods used by researchers, namely 100% and Kappa 1,00 categorized as Excellent. So the conclution is that the K-NN method with K-5 which produces the greatest accuracy and Kappa can be recommended to PKM stakeholders in funding feasibility algorithms.

Fast Multi-Physics Simulation of Microwave Filters via Deep Hybrid Neural Network

One fundamental difficulty in multiphysics numerical simulation is the complex interactions between different physics domains leading to plenty of computational costs. Although neural networks have recently been introduced in multiphysics simulations, the modeling complexity and the enormous amount of training data required may still pose significant challenges to researchers. In this work, we introduce a low-cost, electromagnetic-centric, multiphysics modeling approach to simulate microwave filters. With ground-truth datasets being generated from the finite element method, a novel deep hybrid neural network (DHNN) model structure is introduced, which uses the sigmoid and the ReLU functions as activators to mimic the diversity of biological neurons. A new, more feasible training algorithm is proposed for the efficient development of the DHNN model. The algorithm adopts the design-of-experiment (DOE) sampling technique and is specifically designed for the simulation of multiphysics responses. The strong approximation ability of the DHNN can lead to high-accuracy modeling with fewer training data and less resource consumption. Another advantage of this approach is that the modeling process is more concise and easier to apply compared with other modeling technologies. Numerical examples show that the DHNN can achieve higher accurate results with much less training data compared to traditional ANNs. The advantages of the proposed method in computational efficiency are more pronounced, especially when the amount of input data increases.

Coordinate Descent on the Orthogonal Group for Recurrent Neural Network Training

We address the poor scalability of learning algorithms for orthogonal recurrent neural networks via the use of stochastic coordinate descent on the orthogonal group, leading to a cost per iteration that increases linearly with the number of recurrent states. This contrasts with the cubic dependency of typical feasible algorithms such as stochastic Riemannian gradient descent, which prohibits the use of big network architectures. Coordinate descent rotates successively two columns of the recurrent matrix. When the coordinate (i.e., indices of rotated columns) is selected uniformly at random at each iteration, we prove convergence of the algorithm under standard assumptions on the loss function, stepsize and minibatch noise. In addition, we numerically show that the Riemannian gradient has an approximately sparse structure. Leveraging this observation, we propose a variant of our proposed algorithm that relies on the Gauss-Southwell coordinate selection rule. Experiments on a benchmark recurrent neural network training problem show that the proposed approach is a very promising step towards the training of orthogonal recurrent neural networks with big architectures.

A Weighted-path Following Interior-point Algorithm for Convex Quadratic Optimization Based on Modified Search Directions

Getting a perfectly centered initial point for feasible path-following interior-point algorithms is a hard practical task. Therefore, it is worth to analyze other cases when the starting point is not necessarily centered. In this paper, we propose a short-step weighted-path following interior-point algorithm (IPA) for solving convex quadratic optimization (CQO). The latter is based on a modified search direction which is obtained by the technique of algebraically equivalent transformation (AET) introduced by a new univariate function to the Newton system which defines the weighted-path. At each iteration, the algorithm uses only full-Newton steps and the strategy of the central-path for tracing approximately the weighted-path. We show that the algorithm is well-defined and converges locally quadratically to an optimal solution of CQO. Moreover, we obtain the currently best known iteration bound, namely, $\mathcal{O}\left(\sqrt{n}\log \dfrac{n}{\epsilon}\right)$ which is as good as the bound for linear optimization analogue. Some numerical results are given to evaluate the efffficiency of the algorithm.