Class Of Evolutionary Algorithms Research Articles

An agent-based cooperative co-evolutionary framework for optimizing the production planning of energy supply chains under uncertainty scenarios

Nowadays, energy and power companies compete to get the raw materials and equipment they need on time, as project times lengthen, costs spiral, stock-out continues to plague plans to a decarbonized energy future. The risks reflect the impact of uncertainty and volatility on the resilience of the supply chains. Therefore, there is a need for the enhancement of the production planning in Energy Supply Chains (ESCs), as it enables affordable energy supplies and supports the companies transition to a clean, secure and sustainable energy mix. This study aims to understand the interactive behavior among individuals and optimize their production planning under uncertainty scenarios. In particular, we propose a novel framework to couple an Agent-based Modelling (ABM) and a Co-evolutionary Algorithm (CEA), to realize its capacity to solve a Many-objective Optimization Problem (MaOP) where the profits of multiple agents are concurrently maximized in their interactive transaction processes under normal conditions and uncertain disruption events.For demonstration, we illustrate the proposed approach by considering a five-layer oil and gas ESC model, where uncertainties from multiple sources and the structural dynamics challenge the balance between supply and demand. The results obtained by an integration of a Cooperative Co-evolutionary Particle Swarm Optimizer (CCPSO) algorithm into ABM show the pricing and orders of the target agents are optimized while the loss of ESC resilience is minimized under uncertainty scenarios, proving its capacity of improving the diversity and the convergence, compared to the classic evolutionary algorithms.

A comprehensive cutting tool adaptation method for multi-feature part based on evolutionary game algorithm

The adaptation of cutting tools is the key link in machining. Although a large amount of work has been done to achieve tool selection for a single feature, there is relatively little research on tool scheme adaptation for multi-feature parts. Considering the factors of matching, efficiency, cost, and environment, a comprehensive optimization method for a multi-feature part tool scheme is proposed in this work. First, the critical tool of a complex feature is defined, and a tool combination method with complex single features is proposed to generate an efficient tool strategy. Then, a tool scheme optimization model for multi-feature parts is constructed in which each machining feature is mapped to a game player in the model space, and the set of available tools is mapped to the player strategies. In addition, the classical evolutionary game algorithm is improved to adapt to the model, according to the difference in the features number and type of different parts. Finally, a square test piece is taken as a case to verify the proposed method. The result shows that the method presented in this work can efficiently obtain the customer-preferred tool scheme for the part.

Bio-Inspired Optimization Algorithm Associated with Reinforcement Learning for Multi-Objective Operating Planning in Radioactive Environment.

This paper aims to solve the multi-objective operating planning problem in the radioactive environment. First, a more complicated radiation dose model is constructed, considering difficulty levels at each operating point. Based on this model, the multi-objective operating planning problem is converted to a variant traveling salesman problem (VTSP). Second, with respect to this issue, a novel combinatorial algorithm framework, namely hyper-parameter adaptive genetic algorithm (HPAGA), integrating bio-inspired optimization with reinforcement learning, is proposed, which allows for adaptive adjustment of the hyperparameters of GA so as to obtain optimal solutions efficiently. Third, comparative studies demonstrate the superior performance of the proposed HPAGA against classical evolutionary algorithms for various TSP instances. Additionally, a case study in the simulated radioactive environment implies the potential application of HPAGA in the future.

Multi-subswarm cooperative particle swarm optimization algorithm and its application

Particle swarm optimization (PSO) is a classical evolutionary algorithm and has been widely used to solve continuous optimization problems. However, the performance of the PSO algorithm is highly sensitive to its control parameters and learning strategies. To address this problem, a multi-subswarm cooperative particle swarm optimization algorithm (MSC-PSO) is proposed in this paper. In MSC-PSO, firstly, a population muti-subswarm division method is designed to enhance the global exploration capability. Secondly, a level-based particle adaptive inertia weight strategy is introduced to adjust control parameters. Finally, a level-based update learning mechanism is used for particles adaptive selection learning strategy. The performance of the MSC-PSO is evaluated by the CEC2020 test suite, and five heuristic algorithms are compared with the MSC-PSO in the experiments. The experimental results demonstrate that the proposed MSC-PSO algorithm has significant advantages in terms of convergence speed and solving optimal solutions. Furthermore, the proposed MSC-PSO algorithm is used to solve the plant protection unmanned aerial vehicles (UAV) path planning problem. The experimental results proved that the proposed MSC-PSO algorithm effectively solves the real-world UAV path planning problem, and achieves at most a 34 % reduction in round-trip distance and a 24.1 % reduction in total non-spraying time compared to classical PSO.

A maximum entropy deep reinforcement learning method for sequential well placement optimization using multi-discrete action spaces

Well placement optimization is a crucial method to solve the planar conflicts in reservoir development, mainly to determine the optimal well locations and drilling sequence to maximize the economic benefits of reservoir development. However, the current well placement optimization methods face the problems of high-dimensional discretization of optimization variables and lack of effective incentives for policy exploration, which make it challenging to improve the global optimization ability (the ability to jump out of locally optimal solutions in time and continuously search for better solutions in the whole optimization process) and real-time adjustability of the well placement optimization methods under limited numerical simulation times. In this paper, we propose a new sequential well placement optimization method, based on the Discrete Soft Actor-Critic Algorithm (DSAC), which incorporates the maximum entropy mechanism to formulate well placement and drilling sequencing schemes more efficiently and maximize the net present value (NPV) over the entire life cycle of reservoir development. Specifically, the method models the well placement optimization problem as a Markov Decision Process (MDP) and achieves sequential well placement optimization by training a Deep Reinforcement Learning (DRL) agent that maps reservoir states to a stochastic policy of well placement variables as well as evaluates the value function of the current policy. The DRL agent can determine the optimal infill well location in real-time based on the reservoir state at different times during the development process, thus obtaining the optimal drilling sequence. The proposed method in this paper has two innovations. First, by reconstructing the large-scale discrete action space of well placement optimization variables into multi-discrete action spaces, and with the maximum entropy mechanism, policy exploration is encouraged to improve the global optimization capability. Second, the trained policy can swiftly adapt the subsequent well placement scheme for a specific state of the target reservoir without the requirement to initiate training from scratch, which can realize the offline application of the trained policy and has better real-time adjustability. To verify the effectiveness of the proposed method, it is tested in 2D and 3D reservoir models. The results show that DSAC not only outperforms the gradient-based optimization method, classical evolutionary algorithms, and existing reinforcement learning proximal policy optimization (PPO) method in terms of global optimization ability but also shows better real-time adjustability of the trained policy when applied offline.

A deep reinforcement learning algorithm framework for solving multi-objective traveling salesman problem based on feature transformation

As a special type of multi-objective combinatorial optimization problems (MOCOPs), the multi-objective traveling salesman problem (MOTSP) plays an important role in practical fields such as transportation and robot control. However, due to the complexity of its solution space and the conflicts between different objectives, it is difficult to obtain satisfactory solutions in a short time. This paper proposes an end-to-end algorithm framework for solving MOTSP based on deep reinforcement learning (DRL). By decomposing strategies, solving MOTSP is transformed into solving multiple single-objective optimization subproblems. Through linear transformation, the features of the MOTSP are combined with the weights of the objective function. Subsequently, a modified graph pointer network (GPN) model is used to solve the decomposed subproblems. Compared with the previous DRL model, the proposed algorithm can solve all the subproblems using only one model without adding weight information as input features. Furthermore, our algorithm can output a corresponding solution for each weight, which increases the diversity of solutions. In order to verify the performance of our proposed algorithm, it is compared with four classical evolutionary algorithms and two DRL algorithms on several MOTSP instances. The comparison shows that our proposed algorithm outperforms the compared algorithms both in terms of training time and the quality of the resulting solutions.

A Crisscross-Strategy-Boosted Water Flow Optimizer for Global Optimization and Oil Reservoir Production.

The growing intricacies in engineering, energy, and geology pose substantial challenges for decision makers, demanding efficient solutions for real-world production. The water flow optimizer (WFO) is an advanced metaheuristic algorithm proposed in 2021, but it still faces the challenge of falling into local optima. In order to adapt WFO more effectively to specific domains and address optimization problems more efficiently, this paper introduces an enhanced water flow optimizer (CCWFO) designed to enhance the convergence speed and accuracy of the algorithm by integrating a cross-search strategy. Comparative experiments, conducted on the CEC2017 benchmarks, illustrate the superior global optimization capability of CCWFO compared to other metaheuristic algorithms. The application of CCWFO to the production optimization of a three-channel reservoir model is explored, with a specific focus on a comparative analysis against several classical evolutionary algorithms. The experimental findings reveal that CCWFO achieves a higher net present value (NPV) within the same limited number of evaluations, establishing itself as a compelling alternative for reservoir production optimization.

Thefittest: evolutionary machine learning in Python

Thefittest is a new Python library specializing in evolutionary optimization methods and machine learning methods that use evolutionary optimization methods. Thefittest provides both classical evolutionary algorithms and efficient modifications of these algorithms that do not have implementations in the open access. Among the advantages of the library are the performance of the implemented methods, accessibility and ease of use. The paper discusses the motivation for developing and leading the project, describes the structure of the library with examples of use, and provides a comparison with other projects with a similar development goal. Thefittest is an open-source project published on GitHub and PyPi, developed using modern methods of code analysis and testing. At the moment of writing the paper, the latest version of the library is 0.2.3.

Leukocytes Classification for Leukemia Detection Using Quantum Inspired Deep Feature Selection.

Leukocytes, also referred to as white blood cells (WBCs), are a crucial component of the human immune system. Abnormal proliferation of leukocytes in the bone marrow leads to leukemia, a fatal blood cancer. Classification of various subtypes of WBCs is an important step in the diagnosis of leukemia. The method of automated classification of WBCs using deep convolutional neural networks is promising to achieve a significant level of accuracy, but suffers from high computational costs due to very large feature sets. Dimensionality reduction through intelligent feature selection is essential to improve the model performance with reduced computational complexity. This work proposed an improved pipeline for subtype classification of WBCs that relies on transfer learning for feature extraction using deep neural networks, followed by a wrapper feature selection approach based on a customized quantum-inspired evolutionary algorithm (QIEA). This algorithm, inspired by the principles of quantum physics, outperforms classical evolutionary algorithms in the exploration of search space. The reduced feature vector obtained from QIEA was then classified with multiple baseline classifiers. In order to validate the proposed methodology, a public dataset of 5000 images of five subtypes of WBCs was used. The proposed system achieves a classification accuracy of about 99% with a reduction of 90% in the size of the feature vector. The proposed feature selection method also shows a better convergence performance as compared to the classical genetic algorithm and a comparable performance to several existing works.

A Coevolutionary Algorithm for Many-Objective Optimization Problems with Independent and Harmonious Objectives

Evolutionary algorithm is an effective strategy for solving many-objective optimization problems. At present, most evolutionary many-objective algorithms are designed for solving many-objective optimization problems where the objectives conflict with each other. In some cases, however, the objectives are not always in conflict. It consists of multiple independent objective subsets and the relationship between objectives is unknown in advance. The classical evolutionary many-objective algorithms may not be able to effectively solve such problems. Accordingly, we propose an objective set decomposition strategy based on the partial set covering model. It decomposes the objectives into a collection of objective subsets to preserve the nondominance relationship as much as possible. An optimization subproblem is defined on each objective subset. A coevolutionary algorithm is presented to optimize all subproblems simultaneously, in which a nondominance ranking is presented to interact information among these sub-populations. The proposed algorithm is compared with five popular many-objective evolutionary algorithms and four objective set decomposition based evolutionary algorithms on a series of test problems. Numerical experiments demonstrate that the proposed algorithm can achieve promising results for the many-objective optimization problems with independent and harmonious objectives.

Distributed Evolution Strategies for Black-Box Stochastic Optimization

This work concerns the evolutionary approaches to distributed stochastic black-box optimization, in which each worker can individually solve an approximation of the problem with nature-inspired algorithms. We propose a distributed evolution strategy (DES) algorithm grounded on a proper modification to evolution strategies, a family of classic evolutionary algorithms, as well as a careful combination with existing distributed frameworks. On smooth and nonconvex landscapes, DES has a convergence rate competitive to existing zeroth-order methods, and can exploit the sparsity, if applicable, to match the rate of first-order methods. The DES method uses a Gaussian probability model to guide the search and avoids the numerical issue resulted from finite-difference techniques in existing zeroth-order methods. The DES method is also fully adaptive to the problem landscape, as its convergence is guaranteed with any parameter setting. We further propose two alternative sampling schemes which significantly improve the sampling efficiency while leading to similar performance. Simulation studies on several machine learning problems suggest that the proposed methods show much promise in reducing the convergence time and improving the robustness to parameter settings.

Optimization Methods for Redundancy Allocation in Hybrid Structure Large Binary Systems

This paper addresses the issue of optimal redundancy allocation in hybrid structure large binary systems. Two aspects of optimization are considered: (1) maximizing the reliability of the system under the cost constraint, and (2) obtaining the necessary reliability at a minimum cost. The complex binary system considered in this work is composed of many subsystems with redundant structure. To cover most of the cases encountered in practice, the following kinds of redundancy are considered: active redundancy, passive redundancy, hybrid standby redundancy with a hot or warm reserve and possibly other cold ones, triple modular redundancy (TMR) structure with control facilities and cold spare components, static redundancy: triple modular redundancy or 5-modular redundancy (5MR), TMR/Simplex with cold standby redundancy, and TMR/Duplex with cold standby redundancy. A classic evolutionary algorithm highlights the complexity of this optimization problem. To master the complexity of this problem, two fundamentally different optimization methods are proposed: an improved evolutionary algorithm and a zero-one integer programming formulation. To speed up the search process, a lower bound is determined first. The paper highlights the difficulty of these optimization problems for large systems and, based on numerical results, shows the effectiveness of zero-one integer programming.

Efficient decoupling-assisted evolutionary/metaheuristic framework for expensive reliability-based design optimization problems

Reliability-based design optimization (RBDO) algorithm is to minimize the objective under the probabilistic factors. While gradient-based and classical evolutionary RBDO algorithms provide promising performance on simple optimization problems, they are likely to perform poorly on challenging problems, including the multimodal functions, discrete design spaces, non-differential problems, etc. This paper proposes a unified framework to improve the performance of existing RBDO algorithms for complex RBDO problems. Our framework is based on three new strategies: generalized decoupling evolutionary and metaheuristic RBDO framework, particle’s memory saving strategy, and adaptive fractional-order equilibrium optimizer algorithm. The proposed algorithm is characterized by a decoupling strategy to enable the parallel operation of the inner reliability computation and outer deterministic optimization, a particle’s memory saving strategy to provide effective guidance from the previous iteration, and the adaptive fractional-order equilibrium optimizer algorithm to enhance the search efficiency and global convergence capacity. To evaluate the performance of the proposed algorithm, a wide range of experiments are conducted on different types of use cases. The experimental results demonstrate that our algorithm provides superior performance over other comparative algorithms.

Optimizing Multi-Objective Federated Learning on Non-IID Data with Improved NSGA-III and Hierarchical Clustering

Federated learning (FL) can tackle the problem of data silos of asymmetric information and privacy leakage; however, it still has shortcomings, such as data heterogeneity, high communication cost and uneven distribution of performance. To overcome these issues and achieve parameter optimization of FL on non-Independent Identically Distributed (non-IID) data, a multi-objective FL parameter optimization method based on hierarchical clustering and the third-generation non-dominated sorted genetic algorithm III (NSGA-III) algorithm is proposed, which aims to simultaneously minimize the global model error rate, global model accuracy distribution variance and communication cost. The introduction of a hierarchical clustering algorithm on non-IID data can accelerate convergence so that FL can employ an evolutionary algorithm with a low FL client participation ratio, reducing the overall communication cost of the NSGA-III algorithm. Meanwhile, the NSGA-III algorithm, with fast greedy initialization and a strategy of discarding low-quality individuals (named NSGA-III-FD), is proposed to improve the convergence efficiency and the quality of Pareto-optimal solutions. Under two non-IID data settings, the CNN experiments on both MNIST and CIFAR-10 datasets show that our approach can obtain better Pareto-optimal solutions than classical evolutionary algorithms, and the selected solutions with an optimized model can achieve better multi-objective equilibrium than the standard federated averaging (FedAvg) algorithm and the Clustering-based FedAvg algorithm.

Hierarchical Collaborated Fireworks Algorithm

The fireworks algorithm (FWA) has achieved significant global optimization ability by organizing multiple simultaneous local searches. By dynamically decomposing the target problem and handling each one with a sub-population, it has presented distinct property and applicability compared with traditional evolutionary algorithms. In this paper, we extend the theoretical model of fireworks algorithm based on search space partition to obtain a hierarchical collaboration model. It maintains both multiple local fireworks for local exploitation and one global firework for overall population distribution control. The implemented hierarchical collaborated fireworks algorithm is able to combine the advantages of both classic evolutionary algorithms and fireworks algorithms. Several experiments are provided for in-depth analysis and discussion on the proposed algorithm. The effectiveness of proposed strategy is demonstrated on the benchmark test suite from CEC 2020. Experimental results validate that the hierarchical collaborated fireworks algorithm outperforms former fireworks algorithms significantly and achieves similar results compared with state-of-the-art evolutionary algorithms.

A Rigorous Runtime Analysis of the $$(1 + (\lambda , \lambda ))$$ GA on Jump Functions

The $(1 + (\lambda,\lambda))$ genetic algorithm is a younger evolutionary algorithm trying to profit also from inferior solutions. Rigorous runtime analyses on unimodal fitness functions showed that it can indeed be faster than classical evolutionary algorithms, though on these simple problems the gains were only moderate. In this work, we conduct the first runtime analysis of this algorithm on a multimodal problem class, the jump functions benchmark. We show that with the right parameters, the \ollga optimizes any jump function with jump size $2 \le k \le n/4$ in expected time $O(n^{(k+1)/2} e^{O(k)} k^{-k/2})$, which significantly and already for constant~$k$ outperforms standard mutation-based algorithms with their $\Theta(n^k)$ runtime and standard crossover-based algorithms with their $\tilde{O}(n^{k-1})$ runtime guarantee. For the isolated problem of leaving the local optimum of jump functions, we determine provably optimal parameters that lead to a runtime of $(n/k)^{k/2} e^{\Theta(k)}$. This suggests some general advice on how to set the parameters of the \ollga, which might ease the further use of this algorithm.

Multi-objective path planning for mobile robot with an improved artificial bee colony algorithm.

Effective path planning (PP) is the basis of autonomous navigation for mobile robots. Since the PP is an NP-hard problem, intelligent optimization algorithms have become a popular option to solve this problem. As a classic evolutionary algorithm, the artificial bee colony (ABC) algorithm has been applied to solve numerous realistic optimization problems. In this study, we propose an improved artificial bee colony algorithm (IMO-ABC) to deal with the multi-objective PP problem for a mobile robot. Path length and path safety were optimized as two objectives. Considering the complexity of the multi-objective PP problem, a well-environment model and a path encoding method are designed to make solutions feasible. In addition, a hybrid initialization strategy is applied to generate efficient feasible solutions. Subsequently, path-shortening and path-crossing operators are developed and embedded in the IMO-ABC algorithm. Meanwhile, a variable neighborhood local search strategy and a global search strategy, which could enhance exploitation and exploration, respectively, are proposed. Finally, representative maps including a real environment map are employed for simulation tests. The effectiveness of the proposed strategies is verified through numerous comparisons and statistical analyses. Simulation results show that the proposed IMO-ABC yields better solutions with respect to hypervolume and set coverage metrics for the later decision-maker.

Verification of Classification Model and Dendritic Neuron Model Based on Machine Learning

Artificial neural networks have achieved a great success in simulating the information processing mechanism and process of neuron supervised learning, such as classification. However, traditional artificial neurons still have many problems such as slow and difficult training. This paper proposes a new dendrite neuron model (DNM), which combines metaheuristic algorithm and dendrite neuron model effectively. Eight learning algorithms including traditional backpropagation, classic evolutionary algorithms such as biogeography‐based optimization, particle swarm optimization, genetic algorithm, population‐based incremental learning, competitive swarm optimization, differential evolution, and state‐of‐the‐art jSO algorithm are used for training of dendritic neuron model. The optimal combination of user‐defined parameters of model has been systemically investigated, and four different datasets involving classification problem are investigated using proposed DNM. Compared with common machine learning methods such as decision tree, support vector machine, k‐nearest neighbor, and artificial neural networks, dendritic neuron model trained by biogeography‐based optimization has significant advantages. It has the characteristics of simple structure and low cost and can be used as a neuron model to solve practical problems with a high precision.

The Univariate Marginal Distribution Algorithm Copes Well with Deception and Epistasis.

In their recent work, Lehre and Nguyen (2019) show that the univariate marginal distribution algorithm (UMDA) needs time exponential in the parent populations size to optimize the DeceptiveLeadingBlocks (DLB) problem. They conclude from this result that univariate EDAs have difficulties with deception and epistasis. In this work, we show that this negative finding is caused by the choice of the parameters of the UMDA. When the population sizes are chosen large enough to prevent genetic drift, then the UMDA optimizes the DLB problem with high probability with at most λ(n2+2elnn) fitness evaluations. Since an offspring population size λ of order nlogn can prevent genetic drift, the UMDA can solve the DLB problem with O(n2logn) fitness evaluations. In contrast, for classic evolutionary algorithms no better runtime guarantee than O(n3) is known (which we prove to be tight for the (1+1) EA), so our result rather suggests that the UMDA can cope well with deception and epistatis. From a broader perspective, our result shows that the UMDA can cope better with local optima than many classic evolutionary algorithms; such a result was previously known only for the compact genetic algorithm. Together with the lower bound of Lehre and Nguyen, our result for the first time rigorously proves that running EDAs in the regime with genetic drift can lead to drastic performance losses.

Maximizing Drift Is Not Optimal for Solving OneMax.

It seems very intuitive that for the maximization of the OneMax problem Om(x):=∑i=1nxi the best that an elitist unary unbiased search algorithm can do is to store a best so far solution, and to modify it with the operator that yields the best possible expected progress in function value. This assumption has been implicitly used in several empirical works. In Doerr et al. (2020), it was formally proven that this approach is indeed almost optimal. In this work, we prove that drift maximization is not optimal. More precisely, we show that for most fitness levels between n/2 and 2n/3 the optimal mutation strengths are larger than the drift-maximizing ones. This implies that the optimal RLS is more risk-affine than the variant maximizing the stepwise expected progress. We show similar results for the mutation rates of the classic (1+1) Evolutionary Algorithm (EA) and its resampling variant, the (1+1) EA>0. As a result of independent interest we show that the optimal mutation strengths, unlike the drift-maximizing ones, can be even.