Evolutionary Computation Research Articles

Genetic Programming-Based Evolutionary Deep Learning for Data-Efficient Image Classification

Data-efficient image classification is a challenging task that aims to solve image classification using small training data. Neural network-based deep learning methods are effective for image classification, but they typically require large-scale training data and have major limitations such as requiring expertise to design network architectures and having poor interpretability. Evolutionary deep learning is a recent hot topic that combines evolutionary computation with deep learning. However, most evolutionary deep learning methods focus on evolving architectures of neural networks, which still suffer from limitations such as poor interpretability. To address this, this paper proposes a new genetic programming-based evolutionary deep learning approach to data-efficient image classification. The new approach can automatically evolve variable-length models using many important operators from both image and classification domains. It can learn different types of image features from colour or gray-scale images, and construct effective and diverse ensembles for image classification. A flexible multi-layer representation enables the new approach to automatically construct shallow or deep models/trees for different tasks and perform effective transformations on the input data via multiple internal nodes. The new approach is applied to solve five image classification tasks with different training set sizes. The results show that it achieves better performance in most cases than deep learning methods for data-efficient image classification. A deep analysis shows that the new approach has good convergence and evolves models with high interpretability, different lengths/sizes/shapes, and good transferability.

A Random Particle Swarm Optimization Based on Cosine Similarity for Global Optimization and Classification Problems.

In today's fast-paced and ever-changing environment, the need for algorithms with enhanced global optimization capability has become increasingly crucial due to the emergence of a wide range of optimization problems. To tackle this issue, we present a new algorithm called Random Particle Swarm Optimization (RPSO) based on cosine similarity. RPSO is evaluated using both the IEEE Congress on Evolutionary Computation (CEC) 2022 test dataset and Convolutional Neural Network (CNN) classification experiments. The RPSO algorithm builds upon the traditional PSO algorithm by incorporating several key enhancements. Firstly, the parameter selection is adapted and a mechanism called Random Contrastive Interaction (RCI) is introduced. This mechanism fosters information exchange among particles, thereby improving the ability of the algorithm to explore the search space more effectively. Secondly, quadratic interpolation (QI) is incorporated to boost the local search efficiency of the algorithm. RPSO utilizes cosine similarity for the selection of both QI and RCI, dynamically updating population information to steer the algorithm towards optimal solutions. In the evaluation using the CEC 2022 test dataset, RPSO is compared with recent variations of Particle Swarm Optimization (PSO) and top algorithms in the CEC community. The results highlight the strong competitiveness and advantages of RPSO, validating its effectiveness in tackling global optimization tasks. Additionally, in the classification experiments with optimizing CNNs for medical images, RPSO demonstrated stability and accuracy comparable to other algorithms and variants. This further confirms the value and utility of RPSO in improving the performance of CNN classification tasks.

Modelling Time-dependent Convective Penetration in 1D Stellar Evolution

One-dimensional stellar evolution calculations produce uncertain predictions for quantities like the age, core mass, core compactness, and nucleosynthetic yields; a key source of uncertainty is the modeling of interfaces between regions that are convectively stable and those that are not. Theoretical and numerical work has demonstrated that there should be numerous processes adjacent to the convective boundary that induce chemical and angular momentum transport, as well as modify the thermal structure of the star. One such process is called convective penetration, wherein vigorous convection extends beyond the nominal convective boundary and alters both the composition and thermal structure. In this work, we incorporate the process of convective penetration in stellar evolution calculations using the stellar evolution software instrument mesa. We implement convective penetration according to the description presented by Anders et al. to to calculate a grid of models from the pre-main sequence to helium core depletion. The extent of the convective penetration zone is self-consistently calculated at each time step without introducing new free parameters. We find both a substantial penetration zone in all models with a convective core and observable differences to global stellar properties such as the luminosity and radius. We present how the predicted radial extent of the penetration zone scales with the total stellar mass, age, and metallicity of the star. We discuss our results in the context of existing numerical and observational studies.

Recommendation system using bio-inspired algorithms for urban orchards

According to the Food and Agriculture Organization of the United Nations (FAO), climate change is exponentially affecting agricultural production worldwide, with food prices expected to increase by up to 90 percent by 2030 and hunger and malnutrition rates to rise by 2050. This paper presents the development of a platform based on the Internet of Things (IoT) for monitoring urban gardens as a strategy to mitigate hunger, promote food sovereignty and circular economy in areas of food shortage. To this end, an Internet of Things (IoT) architecture is proposed and implemented that involves a social design layer that allows an effective transfer of knowledge to communities and a recommendation system based on evolutionary computation to optimize and maximize the productivity of urban orchards, and thus contribute to the 2030 agenda of the Sustainable Development Goals (SDGs). Finally, three experiments in urban gardens are shown to validate evolutionary computation and artificial intelligence models, such as multiple linear regression, genetic algorithms, ant colony algorithms and spatial estimation and inference algorithms such as the Kriging algorithm. The productivity of urban lettuce orchards is increased between 25 and 45%.

Brownian motion based multi-objective particle swarm optimization methodology and application in binary classification

The particle swam optimization (PSO) method has been widely applied in evolutionary computation and related industrial areas. However, the optimization performance of traditional PSO is usually influenced by particle updating displacement and velocity, which is randomly distributed and makes the solution solving process inefficient. To deal with this challenge, a Brownian particle motion and Euler-Maruyama (EM) methodology-based model is proposed to improve the performance of the multi-objective PSO. The EM based PSO principles and algorithm framework are introduced by theoretical analysis and mathematical modeling, and then the optimization performance of four comparing PSO models is tested, while the feasibility of the new PSO4 is verified by typical benchmark problems. Afterwards, the characterizations of the EM based PSO method are investigated from the perspective of particle motion mechanism analysis, binary classification application and various optimization performance comparison. The result manifests that the new PSO4 features higher accuracy and less running time among 24 related optimization methods for 4 typical standard datasets of Ionosphere, Breast, Diabetes and Wine. Finally, the multi-objective optimization and classification application potential of the Brownian motion and EM based PSO4 is demonstrated, and the future work regarding the new PSO model is discussed.

MorphVAE: Advancing Morphological Design of Voxel-Based Soft Robots with Variational Autoencoders

Soft robot design is an intricate field with unique challenges due to its complex and vast search space. In the past literature, evolutionary computation algorithms, including novel probabilistic generative models (PGMs), have shown potential in this realm. However, these methods are sample inefficient and predominantly focus on rigid robots in locomotion tasks, which limit their performance and application in robot design automation. In this work, we propose MorphVAE, an innovative PGM that incorporates a multi-task training scheme and a meticulously crafted sampling technique termed ``continuous natural selection'', aimed at bolstering sample efficiency. This method empowers us to gain insights from assessed samples across diverse tasks and temporal evolutionary stages, while simultaneously maintaining a delicate balance between optimization efficiency and biodiversity. Through extensive experiments in various locomotion and manipulation tasks, we substantiate the efficiency of MorphVAE in generating high-performing and diverse designs, surpassing the performance of competitive baselines.

Diversity-guided particle swarm optimization with multi-level learning strategy

Particle swarm optimization (termed as PSO) is a metaheuristic algorithm inspired by the swarm intelligence. Since its advent, PSO has been successfully applied to tackle various issues that are hard to optimize. Unfortunately, similar to other evolutionary computation, PSO is also difficult to get rid of the bad luck of premature convergence and local optimization, especially when dealing with complex multimodal issues. Hence, this work proposes a diversity-guided PSO with multi-level learning strategy (DPSO-MLS). To begin with, chaotic opposition-based learning (OBL) is used to generate well-distributed initial particles to speed up convergence process of DPSO-MLS. Subsequently, according to the current swarm diversity, the high-layer learning mechanism enables PSO to explore the entire search space from a global perspective via attractive and repulsive strategies, respectively. Next, according to average fitness value of entire swarm, the low-layer learning strategy fine-tunes the particles search from a local perspective to sustain its diversity. To be specific, in the context of the repulsion phase, the regular varying function (RVF) embedded update strategy is used to block potential local optima so as to continue exploring the potential search space when the particle's fitness is less than the average of swarm, otherwise an alternative mutation scheme is utilized to enrich the swarm diversity. Accordingly, the slowly varying function (SVF) embedded update strategy is applied when the particle's fitness is greater than the average of swarm, otherwise the worst-best example learning based mechanism is exploited to update the worst particle to improve the quality of swarm in the context of the attraction phase. To verify the effectiveness of DPSO-MLS, extensive experiments are conducted and the results show that our proposal is superior or highly competitive to several existing PSOs according to the robustness, convergence rate and solution accuracy.

Special Issue on Cutting Edge of Reinforcement Learning and its Hybrid Methods

Since deep Q-networks and AlphaGO by Google DeepMind were proposed, not only a reinforcement learning integrated with deep learning but also its applications have attracted much attention on. ChatGPT, which learns by Proximal Policy Optimization as one of reinforcement learning mechanisms, is an excellent example. With this background, we propose the special issue titled “Cutting Edge of Reinforcement Learning and its Hybrid Methods,” which is the same of the special issue in 2017, to cover the latest progress and trends in reinforcement learning and its hybrid methods (combined with machine learning, neural networks, and evolutionary computation). We received 28 submissions, out of which 12 submissions were peer-reviewed and 16 submissions that were outside the scope of the special issue were excluded. After a thorough review process including comments and suggestions by reviewers, seven submissions were accepted. These included four theory and three application papers, indicating nearly equal contributions in theory and application aspects. The content of the first three papers is related to inverse reinforcement learning, while the fourth one is not directly related to reinforcement learning but it would be related to reinforcement learning in the future. All papers address issues surrounding cutting-edge technology in reinforcement learning. We would like to end by expressing that we hope and believe that this special issue can largely contribute to the development in the field of reinforcement learning while holding a wide international appeal.

Evolutionary Computation: Theories, Techniques, and Applications

Evolutionary computation is now nearly 50 years old, originating with the seminal work of John Holland at the University of Michigan in 1975 which introduced the genetic algorithm [...]

Introduction to the Special Issue on Explainable AI in Evolutionary Computation

Introduction to the Special Issue on Explainable AI in Evolutionary Computation

Hedging policy using neural networks and its combination with heuristic algorithms case study: Dez Reservoir

In recent years, the use of various Artificial Intelligence (AI) methods, such as evolutionary computation, heuristic algorithms, artificial neural networks, and fuzzy theory calculations, has gained popularity in addressing water resources issues. These algorithms have shown great success in solving problems that traditional deterministic methods struggle with. This study focuses on optimizing Dez reservoir operation over a long-term period using a nonlinear loss function through an evolutionary artificial neural network algorithm. The outcomes of this approach are then contrasted with genetic exploration and harmony search algorithms, highlighting the strengths and weaknesses of each method. Ultimately, a combination of the evolutionary artificial neural network method and hedging models is employed for optimal reservoir management, with results compared to the previous approach. Results show the appropriate performance of combining hedging policy with artificial neural network and harmony search algorithm. This combination significantly reduces the vulnerability value with a slight decrease in reliability.

Deep Reinforcement Learning and Auto-Differential Evolution Co-Guided Coal Washing

Background. Coal washing is a complicated process and difficult to control, which has many controlling parameters with strong coupling relationship. It is still a challenge to realize the self-perception, self-adjustment, and self-evaluation of coal washing machine, improve the quality of coal washing, ensure production safety, and reduce labor cost. Methods. Through the intelligent transformation of jig, this paper proposes an intelligent washing method with cooperated deep reinforcement learning and evolutionary computation. First, it designs a fault warning method based on statistical analysis, helping to recover the normal running state of jig with manual maintenance. Then, it constructs a regulation strategy generation method with deep reinforcement learning supported by the fusion of artificial experience and historical data. Last, for the lack of monitoring data caused by poor communication quality and environment, the regulation strategy prediction method with evolutionary computation and surrogate model is proposed. Results. In practice, this method shows accurate fault warning accuracy and rapid cleaned coal ash adjustment response ability. Conclusions. This shows that the method proposed in this paper is of great significance for intelligent washing and can better cope with the special situation when the washing equipment sensing data are missing.

Industrial defect detection and location based on greedy particle swarm optimization algorithm

ABSTRACT Particle swarm optimization algorithm is a branch of evolutionary computing and can search for a better solution in a given feature space. This paper introduces the particle swarm optimization algorithm and greedy strategy for the defect detection and location of industrial components and then proposes a greedy particle swarm optimization algorithm. This paper employs sparse random projection to map the vast high-dimensional information to the low-dimensional space while maintaining the relative distance between the data. This employment helps to accelerate the training and prediction speed of the model and remove some unimportant features or noise. This algorithm first adopts particle swarm optimization (PSO) to initialize the cluster centers in an effort to minimize the maximum distance between all unlabeled data points and these centers. The algorithm then utilizes the greedy strategy to select a batch of data points to represent the corresponding features of the normal image, thereby improving the coverage of the model to the data. Experiments have shown that the results of most categories of data sets are close to or better than the current existing methods, especially in defect detection. In terms of defect localization for object categories, our method achieves a pixel-level anomaly localization index (AUROC) of 98.3% on the MVTec AD dataset.

Route design in sheepdog system–traveling salesman problem formulation and evolutionary computation solution–

ABSTRACT In this study, we consider the guidance control problem of the sheepdog system, which involves the guidance of a flock using the characteristics of the sheepdog and sheep. Sheepdog systems require strategies to guide sheep agents to a goal area using small numbers of sheepdog agents, and various methods have been proposed. Previous studies have proposed a guidance control law to reliably guide sheep herds; however, the movement distance required by the sheepdog for guidance has not been considered. Therefore, in this study, we propose a novel guidance algorithm in which a supposedly efficient route for guiding a flock of sheep is designed via the traveling salesman problem and evolutionary computation. Numerical simulations were performed to confirm whether sheep flocks could be guided and controlled using the obtained guidance routes. We revealed that the proposed method reduces both the guidance failure rate and the guidance distance.

Hyperscale data analysis oriented optimization mechanisms for higher education management systems platforms with evolutionary intelligence

Higher education institutions face vast amounts of complex data from various sources. Extracting meaningful insights from this data can improve student outcomes and institutional performance. This paper proposes novel evolutionary optimization mechanisms for higher education management systems to enable hyperscale data analysis. An integrated framework is developed, combining educational data mining, learning analytics, and evolutionary computation techniques. The methodology employs a multi-objective evolutionary algorithm with dynamic resource allocation to optimize multiple objectives simultaneously. Adaptive learning control is incorporated to balance exploration and exploitation. Theoretical analyses provide convergence proofs for the proposed algorithms. Comprehensive experiments on real-world and synthetic datasets demonstrate the effectiveness of the proposed mechanisms compared to state-of-the-art approaches. The results show significant performance gains regarding solution quality, scalability, and computational efficiency. The proposed techniques can be a foundation for developing the next generation of intelligent higher education management systems.

Near-Earth asteroids of cometary origin associated with the Virginid complex

The Virginid meteoroid streams produce a series of meteor showers active annually during February–May. A certain parent comet is not found but a related association of some showers with near-Earth asteroids was previously established and a cometary origin of these asteroids was suggested. We performed a new search for NEAs belonging to the Virginid asteroid–meteoroid complex. On the base of calculation of orbital evolution of a sample of NEAs and determination of theoretical features of related showers a search for observable active showers close to theoretically predicted ones was carried out. As a result, the predicted showers of 27 NEAs were identified with the showers of the Virginid complex. Revealed association points to a cometary nature of NEAs that are moving within the stream and may be considered as extinct fragments of a larger comet-progenitor of the Virginid asteroid–meteoroid complex.

Toward Robust and Accurate Myoelectric Controller Design Based on Multiobjective Optimization Using Evolutionary Computation

Toward Robust and Accurate Myoelectric Controller Design Based on Multiobjective Optimization Using Evolutionary Computation

Genetic Algorithms and Evolutionary Computing in Python

Genetic Algorithms and Evolutionary Computing in Python

A Novel Evolutionary Algorithm With Column and Sub-Block Local Search for Sudoku Puzzles

Sudoku puzzles are not only popular intellectual games but also NP-hard combinatorial problems related to various real-world applications, which have attracted much attention worldwide. Although many efficient tools, such as evolutionary computation (EC) algorithms, have been proposed for solving Sudoku puzzles, they still face great challenges with regard to hard and large instances of Sudoku puzzles. Therefore, to efficiently solve Sudoku puzzles, this paper proposes a genetic algorithm (GA)-based method with a novel local search technology called local search-based GA (LSGA). The LSGA includes three novel design aspects. First, it adopts a matrix coding scheme to represent individuals and designs the corresponding crossover and mutation operations. Second, a novel local search strategy based on column search and sub-block search is proposed to increase the convergence speed of the GA. Third, an elite population learning mechanism is proposed to let the population evolve by learning the historical optimal solution. Based on the above technologies, LSGA can greatly improve the search ability for solving complex Sudoku puzzles. LSGA is compared with some state-of-the-art algorithms at Sudoku puzzles of different difficulty levels and the results show that LSGA performs well in terms of both convergence speed and success rates on the tested Sudoku puzzle instances.

Simulation of the time-dependent propagation of leak-triggered helium density gradients in cryogenic and nonisothermal accelerator vacuum systems

A novel simulation code for the calculation of the time-dependent evolution of helium density profiles in tubular-shaped beam vacuum systems of particle accelerators is presented. The code called TransVac was written in the statistics programming language R using the noncommercial development software RStudio and is based on an analytical approach. In contrast to earlier simulation tools based on analytical computational methods, the new code does not only master the profile calculation in conventional vacuum systems operated at room temperature, but also in fully cryogenic and in nonisothermal systems composed of cryogenic and warm sections. Data of helium adsorption isotherms are used to calculate gas densities profiles in cold vacuum systems with the cryosorption-based wall-pumping effect. The article discusses how the simulation code works and which mathematical algorithm is used. Comparisons between experimental and theoretical data confirm that the software developed provides sufficiently reliable predictions on the propagation behavior of leak-triggered helium pressure waves in cryogenic and room-temperature vacuum systems as well.