Quantum Evolutionary Algorithm Research Articles

Balanced quantum neural architecture search

In the last decade, there has been continuously increasing attention on Neural Architecture Search (NAS). The design of network architecture is aimed at automatically generating efficient neural networks in the absence of prior knowledge. Most of current oneshot NAS methods are based on weight sharing. However, there are two main problems that lead to suboptimal search results. Firstly, the large scale supernet results in a complex and difficult search for optimal subnetwork structures in the large discrete search space. Secondly, weight sharing makes the internal weight parameters of the supernet cannot always vary towards the final weights required for the optimal network structure due to the influence of different structures. These problems are responsible for the fact that the performance of the subnetworks searched by the supernet cannot represent the real performance of the subnetworks trained from scratch. To solve these problems, we propose the method based on quantum evolution and balance pool, called BQNAS, which consists of two stages. The single path supernet is trained based on weight sharing and balance pool sampling method. Then, quantum parallelism is exploited for one-hot encoding and searching for the optimal subnetwork structure based on quantum evolutionary algorithms. It can greatly improve the search efficiency of oneshot approaches in subnetwork sampling and evaluate the real performance for searching for superior subnetwork structure. Extensive experiments on two datasets show that the proposed approach outperforms the state-of-the-art ones. Specifically, BQNAS achieves a top-1 accuracy of 97.27% on CIFAR-10 and 81.36% on CIFAR-100.

Adjustable rotation gate based quantum evolutionary algorithm for energy optimisation in cloud computing systems

Adjustable rotation gate based quantum evolutionary algorithm for energy optimisation in cloud computing systems

A Scalable Open-Set ECG Identification System Based on Compressed CNNs.

Deep learning (DL) is known for its excellence in feature learning and its ability to deliver high-accuracy results. Its application to ECG biometric recognition has received increasing interest but is also accompanied by several deficiencies. In this study, we focus on applying DL, especially convolutional neural networks (CNNs), to ECG biometric identification to address these deficiencies. Using prestored user-specific feature vectors, the proposed scheme can exclude unregistered subjects to realize "open-set" identification. With the help of its scalable structure and "transfer learning," new subjects can be enrolled in an existing system without the need for storing the ECGs of those previously enrolled. Finally, schemes based on the quantum evolutionary algorithm (QEA) are presented to prune unnecessary filters in the proposed CNN model. The performance of the proposed scheme was evaluated using the ECGs of 285 subjects from the PTB dataset. The experimental results demonstrate an identification rate of more than 99% in closed-set identification. Although incorporating the proposed method for unregistered subject exclusion degraded the identification performance slightly, the ability of the approach to resist a dictionary attack was evident. Finally, using the QEA-based filter pruning method and its two-stage extension reduced the number of floating-point operations required to complete one identity recognition to 1.20% and 0.22% of the original value without significantly impacting the identification accuracy.

Computer-Aided Clustering Analysis of Short-Term Interactive Data of Industry Microblog Marketing Effect and Number of Fans Based on Quantum Evolutionary Game Algorithm

Computer-Aided Clustering Analysis of Short-Term Interactive Data of Industry Microblog Marketing Effect and Number of Fans Based on Quantum Evolutionary Game Algorithm

Joint optimization of low-carbon product family configuration and smart production line selection

To meet the requirements of small batch orders and rapid production, many enterprises began to use robots, artificial intelligence, cloud computing, and other new technologies to construct smart production lines, which are high-flexible and capable of autonomously diagnosing their health and autonomously designing continuous improvement projects. In this study, a joint optimization model is established for concurrently determining the optimal product family configuration and the optimal construction plan for a smart production line. Greenhouse gas emissions are considered in the joint optimization model to realize the low-carbon design concept. A quantum evolutionary algorithm and a genetic algorithm are developed to solve the established optimization model. A case study of the electronic dictionary product family is provided to illustrate the proposed method. The numeric experiments show that the proposed model can reap more profit from a systematic perspective compared with the traditional two-stage approach, and the proposed quantum evolutionary algorithm has a good performance in solving the joint optimization problem.

Size and Topology Optimization of Truss Structures Using Quantum-Based HS Algorithm

With the development of various metaheuristic algorithms, research cases that perform weight optimization of truss structures are steadily progressing. In particular, due to the possibility of developing quantum computers, metaheuristic algorithms combined with quantum computation are being developed. In this paper, the QbHS (Quantum based Harmony Search) algorithm was proposed by combining quantum computation and the conventional HS (Harmony Search) algorithms, and the size and topology optimization of the truss structure was performed. The QbHS algorithm has the same repetitive computational structure as the conventional HS algorithm. However, the QbHS algorithm constructed QHM (Quantum Harmony Memory) using the probability of Q-bit and proposed to perform pitch adjusting using the basic state of Q-bit. To perform weight optimization of truss structures using the proposed QbHS algorithm, 20 bar, 24 bar, and 72-bar truss structures were adopted as examples and compared with the results of the QE (Quantum Evolutionary) algorithm. As a result, it was confirmed that the QbHS algorithm had excellent convergence performance by finding a lower weight than the QE algorithm. In addition, by expressing the weight optimization results of the truss structure with an image coordinate system, the topology of the truss structure could be confirmed only by the picture. The results of this study are expected to play an important role in future computer information systems by combining quantum computation and conventional HS algorithms.

The cutting path planning of main components in China initiative accelerator driven subcritical system based on the quantum evolutionary algorithm

The CiADS is a promising nuclear demonstration facility for the purpose of transmuting long-life fission products and minor actinides to the short-life isotopes. The main components in CiADS are the large volume equipment with complex structure and high radioactivity, which are also the important source items for the decommissioning plan design that should be cut into smaller blocks for the proper management and the safety storage. However, it has been believed as the extremely difficult and challenging task to construct the cutting paths on the main components in order to obtain the optimized blocks with acceptable range of radioactivity, since the cutting paths are hard to describe with explicit expressions in the complex model structures, and the manual selection of cutting blocks with suitable radioactivity is the error prone and labor intensity task. Considering this situation, the cutting path planning method has been developed in this study. During the decommissioning process, the complex models of main components have been converted to the simplified geometries with voxelized description, and then the quantum evolutionary algorithm has been used to automatically find the appropriate cutting blocks by means of the voxelized description for the purpose of meeting the decommissioned requirements. Moreover, the reconstruction procedure has been used to record the topological information of cutting paths on the main components with the visualization function in the graphical user interface. Finally, the verification and validation of the cutting path planning method have been tested in the decommissioning scenario, and all the numerical results show that the optimized method can effectively fulfill the cutting path planning tasks of the CiADS facility.

Adjustable rotation gate-based quantum evolutionary algorithm for energy optimisation in cloud computing systems

Adjustable rotation gate-based quantum evolutionary algorithm for energy optimisation in cloud computing systems

A balanced-quantum inspired evolutionary algorithm for solving disassembly line balancing problem

Disassembly lines are the most efficient choice for recovering parts and components on a large scale from old products. The efficient utilization of disassembly line resources requires optimal selection and sequencing of disassembly tasks which constitutes the disassembly line balancing problem (DLBP). In this work, a novel Balanced Quantum-inspired evolutionary algorithm (Balanced-QEA) is proposed to optimize profit and workload balance for the DLBP. Quantum evolutionary algorithms (QEA) utilize stochastic solution representation in the form of q-bits to explore the solution space. The proposed approach differs from traditional QEA by strategically making multiple observations for a single quantum individual. This modification aims to address the weakness of traditional QEA by utilizing stochastic information in quantum solutions more effectively. The application of the proposed approach is illustrated numerically using an example of radio set to maximize profit and workload balance. For validating the utility of proposed modification, the performance of Balanced-QEA is compared with traditional QEA and five other popular evolutionary algorithms in the field of DLBP. The comparisons are done using benchmark disassembly instances of different scales. Results show that the proposed Balanced QEA is superior to QEA and other algorithms in terms of best solutions.

Product family configuration optimisation considering after-sale service: an adaptive quantum evolutionary algorithm approach

ABSTRACT By providing after-sale services, manufacturing companies can gain more chances to interact with customers, improve brand preference and loyalty, occupy new bases in the value chain, and create more cross-selling opportunities. Considering after-sale service, we propose a new method for product family configuration optimisation in this study. Mathematical models for product family configuration optimisation are established with the objective of maximising the overall profit of the family of products and after-sale services. An adaptive quantum evolutionary algorithm (AQEA) is developed to solve the established optimisation models. An example of an e-book reader product is used to illustrate the proposed method in a case study. An industry case study shows that: (1) the proposed method outperforms the traditional method; (2) the proposed AQEA has better performance when compared with the other three meta-heuristic algorithms; and (3) the proposed method is more useful for industrial cases where the proportion of service profit is relatively large.

Dynamic hybrid mechanism-based differential evolution algorithm and its application

In order to effectively schedule railway train delay, an adaptive cooperative co-evolutionary differential evolution with dynamic hybrid mechanism of the quantum evolutionary algorithm and genetic algorithm, named QGDECC is designed in this paper. In the QGDECC, the quantum variable decomposition strategy is designed by utilizing qubit string to decompose variables adaptively according to the coevolution performance. Then the increment mutation method is proposed to improve the convergence speed which make full use of searched evolution information. Besides, the parameter adaptive strategy is deeply explored for strengthening the robust of the algorithm. The QGDECC with global search capability is employed to realize a railway train delay scheduling method for effectively eliminating the impact of train delay. Finally, several benchmark functions and actual train operation data are selected to verify the optimization performance of QGDECC. The experimental results show that QGDECC has higher adaptability, faster convergence speed and accuracy. The train delay scheduling method can effectively eliminate the impact of delay on the railway network, and minimize the gap between the rescheduled train schedule and the original train schedule.

Interval Prediction Method for Wind Speed Based on ARQEA Optimized by Beta Distribution and SWLSTM

The interval prediction of wind speed is crucial for the economic and safe operation of wind farms. To overcome the probability density function parameter optimization and long-term correlation of time series problems in an interval prediction method, a hybrid model based on the beta distribution of an allele real-coded quantum evolutionary algorithm (ARQEA) and a shared weight long short-term memory (SWLSTM) neural network is proposed for predicting the interval of short-term wind speed, which is beta–ARQEA–SWLSTM. Input variables are determined via autocorrelation functions, and the shape and position parameters in the beta distribution function are optimized by the ARQEA algorithm. An interval-divided multi-distribution function aggregation is proposed to deal with the fluctuation of wind speed series. Lastly, case studies are provided to demonstrate the effectiveness of the proposed method.

Optimization of Interval Type-2 Fuzzy Logic Controller Using Real-Coded Quantum Clonal Selection Algorithm

In recent years, quantum computing has gained immense popularity with the production of real quantum computers. Researchers have developed quantum-inspired evolutionary algorithms (QIEAs) to solve combinatorial optimization problems and have obtained successful results. As a special case of QIEAs, real-coded quantum evolutionary algorithm (RCQEA) is used in the optimization of high-dimensional complex problems. In this study, a novel mechanism of the quantum rotation gate (QRG) that is used to determine the rotation angle of the qubit in the RCQEA is introduced and implemented to accelerate the evolutionary process and increase the possibility of finding the optimal solution. Moreover, the skeleton of RCQEA is modified by using a clonal selection mechanism, and the real-coded quantum clonal selection algorithm (RCQCSA) is developed. Our proposed QRG accelerates the convergence speed of the algorithm. The main purpose of this study is to present a more effective algorithm that inspires quantum computing principles for optimizing the interval type-2 fuzzy logic controller (IT2FLC) membership functions (MFs). In this study, four different comparisons are made with these two different algorithms that have the original version of QRG and our proposed QRG. Optimized IT2FLC provides stabilization of the inverted pendulum system. The results show that the RCQCSA having our proposed QRG outperforms RCQEA in stabilizing the inverted pendulum system by optimizing the IT2FLC parameters.

Power grid expansion planning model based on improved quantum evolutionary algorithm under supply-demand balance constraints

In order to improved the accuracy of power grid expansion planning, shortened the planning time and reduce the investment cost, a power grid expansion planning model based on improved quantum evolutionary algorithm under the constraint of supply and demand balance is proposed. Using SWOT analysis method and analytic hierarchy process, this paper qualitatively analyzes the comprehensive situation of Shaanxi Power Grid Company, establishes the evaluation index of supply-demand relationship, and constructs the power grid expansion planning model. Through the rotation angle strategy, the quantum evolutionary algorithm is improved to solve the power grid expansion planning model and obtain the optimal solution of power grid expansion planning. The experimental results show that the proposed method has shorter planning time and higher accuracy, and can effectively reduce the investment cost.

Comparison of differential evolution, particle swarm optimization, quantum-behaved particle swarm optimization, and quantum evolutionary algorithm for preparation of quantum states

Four intelligent optimization algorithms are compared by searching for control pulses to achieve the preparation of target quantum states for closed and open quantum systems, which include differential evolution (DE), particle swarm optimization (PSO), quantum-behaved particle swarm optimization (QPSO), and quantum evolutionary algorithm (QEA). We compare their control performance and point out their differences. By sampling and learning for uncertain quantum systems, the robustness of control pulses found by these four algorithms is also demonstrated and compared. The resulting research shows that the QPSO nearly outperforms the other three algorithms for all the performance criteria considered. This conclusion provides an important reference for solving complex quantum control problems by optimization algorithms and makes the QPSO be a powerful optimization tool.

A distributed adaptive optimization spiking neural P system for approximately solving combinatorial optimization problems

An optimization spiking neural P system (OSNPS) aims to obtain the approximate solutions of combinatorial optimization problems without the aid of evolutionary operators of evolutionary algorithms or swarm intelligence algorithms. To develop the promising and significant research direction, this paper proposes a distributed adaptive optimization spiking neural P system (DAOSNPS) with a distributed population structure and a new adaptive learning rate considering population diversity. Extensive experiments on knapsack problems show that DAOSNPS gains much better solutions than OSNPS, adaptive optimization spiking neural P system, genetic quantum algorithm and novel quantum evolutionary algorithm. Population diversity and convergence analysis indicate that DAOSNPS achieves a better balance between exploration and exploitation than OSNPS and AOSNPS.

What an “Ehm” Leaks About You: Mapping Fillers into Personality Traits with Quantum Evolutionary Feature Selection Algorithms

This work shows that fillers - short utterances like “ehm” and “uhm” - allow one to predict whether someone is above median along the Big-Five personality traits. The experiments have been performed over a corpus of 2,988 fillers uttered by 120 different speakers in spontaneous conversations. The results show that the prediction accuracies range between 74 and 82 percent depending on the particular trait. The proposed approach includes a feature selection step - based on Quantum Evolutionary Algorithms - that has been used to detect the personality markers, i.e., the subset of the features that better account for the prediction outcomes and, indirectly, for the personality of the speakers. The results show that only a relatively few features tend to be consistently selected, thus acting as reliable personality markers.

Quantum evolutionary algorithm based power optimization control strategy for China initiative accelerator driven subcritical system

The power optimization control strategy for China initiative Accelerator Driven subcritical System (CiADS), which focuses on the study of dynamic features and the stable operation of coupling nuclear system, is one of the most important research topics for the current engineering stage that is expected to be completed in the year 2025. However, the traditional way in handling the power control problem was carried out by specialists with transcendental knowledge and experiences, which has been believed as the high labor intensity and the brute force trial-and-error process. In this context, the Quantum Evolutionary Algorithm (QEA) has been employed to deal with the CiADS power optimization control strategy in order to fulfill the adaptive searching procedure continuously and automatically. The manipulation of proton beam intensity has been used as the main control approach, and the physical model in describing the dynamic characteristics of power control system has been constructed in coupling the lumped parameter method and the point reactor kinetics method. In addition, the logical structure of Proportional Integral Derivative (PID) controller has been established for the power control process, and the QEA has been adopted to obtain the suitable set of parameters in the PID controller. Finally, the performance of the optimization method has been verified in the predefined control scenarios, and all the numerical results show that the control method has the applicability and feasibility in solving the CiADS power control problems.

Cloud Education Chain and Education Quality Evaluation Based on Hybrid Quantum Neural Network Algorithm

This paper proposes the functional model and application service implementation process of the education cloud platform application service architecture. The entire cloud application service architecture mainly includes four parts: cloud service management, cloud application service rapid creation and deployment, dynamic process configuration, and unified identity authentication. Based on the basic theory of workflow, the process status and business services of cloud application services are discussed. The BP neural network weight optimization model based on the improved quantum evolution method is studied, and a method that combines the improved quantum evolution algorithm (IQEA) and the BP algorithm to complete the back propagation neural network training is proposed, that is, the IQEA-BP algorithm. Firstly, the traditional quantum evolution algorithm is improved, and then, the improved quantum evolution algorithm is used to optimize the network weights as a whole to overcome the shortcomings of the BP algorithm that is easy to fall into the local optimum; then, we use the BP algorithm to find the better weight as the initial value to improve the training and prediction accuracy of the network. In order to enrich the school education quality evaluation system, this article adds soft indicators that can reflect school education performance on the basis of the existing “National Education Inspection Team” indicators and uses analytical methods to prove the effectiveness and feasibility of the new evaluation indicators. The X1-X10 index data is selected as the evaluation index of the school education quality evaluation system in this paper. Testing the performance of the BP neural network, the accuracy rate of the school education quality evaluation is 93.3%, the average absolute error is 0.067, and the accuracy and recall rate of the test set grade gradient of 0, 1, 2, 3, 5, 6, and 8 are all 93%, indicating that the IQEA-BP neural network algorithm has a good effect on the evaluation of school education quality.

Bi-Level Multi-Objective Optimization Scheduling for Regional Integrated Energy Systems Based on Quantum Evolutionary Algorithm

Integrated energy systems have become an important research topic in the pursuit of sustainable energy development. This paper examines regional integrated energy systems, presents the typical architecture of regional integrated energy systems, and builds an integrated energy system model. Two evaluation indexes are proposed: the integrated energy self-sufficiency rate and the expected energy deficiency index. Based on these evaluation indexes and taking into account the uncertainty of wind power generation, a bi-level optimization model based on meta-heuristic algorithms and multi-objective programming is established to solve the problem of regional integrated energy system planning under different load structures and for multi-period and multi-scenario operation modes. A quantum evolutionary algorithm is combined with genetic algorithms to solve the problem.