Evolution Strategy Research Articles

A conserved nuclear factor YC subunit, NF-YC3, is essential for arbuscule development.

Establishing reciprocal symbiosis with arbuscular mycorrhizal (AM) fungi is an important evolutionary strategy of most terrestrial plants to adapt to environmental stresses, especially phosphate (Pi) deficiencies. Identifying the key genes essential for AM symbiosis in plants and dissecting their functional mechanisms will be helpful for the breeding of new crop varieties with enhanced nutrient uptake efficiency. Here, we report a nuclear factor YC subunit-encoding gene, OsNF-YC3, whose expression is specifically induced in arbuscule-containing cells, plays an essential role in AM symbiosis. Knockout of OsNF-YC3 resulted in stunted arbuscule morphology and substantially decreased P accumulation, while overexpressing OsNF-YC3 enhanced mycorrhization and Pi uptake efficiency. OsNF-YC3 is directly regulated by OsPHRs, the major regulators of Pi starvation responses. Chromatin immunoprecipitation sequencing analysis uncovered multiple genes with crucial roles in arbuscule development as its potential downstream targets, including the AM-specific Pi transporter gene OsPT11. OsNF-YC3 can form a heterotrimer with the other two NF-Y subunits, OsNF-YA11 and OsNF-YB11, in yeast. Loss of OsNF-YA11 function also severely impaired arbuscule development in its mutants. Overall, our results highlight an essential role of OsNF-YC3 and its potential interacting NF-Y subunit, OsNF-YA11, in regulating AM symbiosis and arbuscule development.

Enhancing neural network applications in phonetic classification through population-based incremental learning

Significant progress has been achieved in the field of Automatic Speech Recognition (ASR) thanks to the development of powerful algorithms. Research based on simplified biological models has led to the emergence of a new class called Estimation of Distribution Algorithms (EDA), which preserves significant partial solutions. Population-Based Incremental Learning (PBIL) is a technique that combines stochastic search and optimization. It is a statistical approach with evolutionary computation similar to EDAs, aimed at adapting recognition systems based on artificial Neural Networks (NN). Our main contribution lies in the improvement achieved by PBIL, which outperforms Genetic Algorithms (GA) by 16% and Evolutionary Strategies (ES) by 9.93%.

Elevational differentiation occurs alongside high plasticity in a general‐purpose genotype invasive plant

Abstract Highly plastic general‐purpose genotypes are a frequent occurrence among invasive plants. Yet it remains uncertain to what extent genetic differentiation can co‐occur with such elevated levels of plasticity. Understanding the interplay between these two evolutionary strategies is essential to understand potential invasive success and future climate change responses. We investigated the potential for genetic shifts between upland and lowland populations of the highly plastic invasive herb Erythranthe guttata in New Zealand. We aimed to determine (a) the presence of genetic differentiation between upland and lowland origin E. guttata populations; (b) whether any differences aligned with established adaptive patterns; and (c) whether genetic differentiation was occurring alongside the high plasticity of E. guttata populations in response to varying elevations. We grew cuttings from 38 E. guttata populations from upland and lowland Canterbury in a lowland and an upland common garden, where we measured a wide range of growth and reproductive traits. We found significant differentiation between upland and lowland origin populations over most measured traits. Plants sourced from upland populations flowered earlier and produced more flowers than plants originating from lowland populations. Lowland origin plants were taller, had larger leaves and higher photosynthetic rates than upland plants. These differences occurred alongside high levels of unspecialised plasticity to the growing environment. Synthesis: We found that over a period of less than 150 years, distinct lowland and upland phenotypes of E. guttata have emerged. These differences are consistent with known selective patterns to elevation that favour reproductive security at higher elevations and competitive ability at lower elevations. This rapid genetic differentiation occurred alongside high plasticity to growing environment, suggesting that highly plastic invasive species still retain the capacity to genetically adapt to novel environments.

EGNAS: Efficient Graph Neural Architecture Search Through Evolutionary Algorithm

The primary objective of our research is to enhance the efficiency and effectiveness of Neural Architecture Search (NAS) with regard to Graph Neural Networks (GNNs). GNNs have emerged as powerful tools for learning from unstructured network data, compensating for several known limitations of Convolutional Neural Networks (CNNs). However, the automatic search for optimal GNN architectures has seen little progressive advancement so far. To address this gap, we introduce the Efficient Graph Neural Architecture Search (EGNAS), a method that leverages the advantages of evolutionary search strategies. EGNAS incorporates inherited parameter sharing, allowing offspring to inherit parameters from their parents, and utilizes half epochs to improve optimization stability. In addition, EGNAS employs a combined evolutionary search, which explores both the model structure and the hyperparameters within a large search space, resulting in improved performance. Our experimental results demonstrate that EGNAS outperforms state-of-the-art methods in node classification tasks on the Cora, Citeseer, and PubMed datasets while maintaining a high degree of computational efficiency. In particular, EGNAS is the fastest GNN architecture search method in terms of search time, particularly when compared to precedently suggested evolutionary search strategies, delivering performance up to 40 times faster.

A prescription for engineering PFAS biodegradation.

Per- and polyfluorinated chemicals (PFAS) are of rising concern due to environmental persistence and emerging evidence of health risks to humans. Environmental persistence is largely attributed to a failure of microbes to degrade PFAS. PFAS recalcitrance has been proposed to result from chemistry, specifically C-F bond strength, or biology, largely negative selection from fluoride toxicity. Given natural evolution has many hurdles, this review advocates for a strategy of laboratory engineering and evolution. Enzymes identified to participate in defluorination reactions have been discovered in all Enzyme Commission classes, providing a palette for metabolic engineering. In vivo PFAS biodegradation will require multiple types of reactions and powerful fluoride mitigation mechanisms to act in concert. The necessary steps are to: (1) engineer bacteria that survive very high, unnatural levels of fluoride, (2) design, evolve, and screen for enzymes that cleave C-F bonds in a broader array of substrates, and (3) create overall physiological conditions that make for positive selective pressure with PFAS substrates.

A decomposition-based dynamic constrained multi-objective task assignment for heterogeneous crowdsensing

Efficient task allocation is a crucial issue of Mobile crowdsensing (MCS). Generally, only homogeneous mobile users like human are selected as the participants, causing a difficulty to meet the spatiotemporal coverage demand on human-unreachable regions. To overcome this drawback, unmanned aerial vehicles are introduced to form heterogeneous MCS, which can be formulated into a dynamic constrained multi-objective task allocation model. Taking the maximum average sensing quality of all tasks and the maximum average remaining budget for each subtask as the optimization objectives, an improved decomposition-based multi-objective evolutionary algorithm is presented to find the optimal allocation scheme. Specifically, the problem is first decomposed into a set of dynamic constrained scalar subproblems. For each subproblem, a stochastic configuration network (SCN)-based initialization is developed to produce the promising population, in which SCNs learn the probabilities of mobile users being allocated to each task. Following that, a reinforcement learning-based autonomous evolutionary strategy is adopted to recommend the most appropriate solvers in terms of the state of current population. A hybrid population update mechanism is then employed to form the high-quality offspring, with the purpose of balancing the feasibility, convergence and diversity. The extensive experiments on 20 dynamic instances are conducted to demonstrate the effectiveness of proposed algorithm compared to other task allocation algorithms.

Information‐Sharing Strategies Based on Trust Risks Under a Governmental Incentive Mechanism

ABSTRACTDue to swift shifts in market demand and variable production capacities within the chip industry, trust risks between upstream and downstream entities can readily develop, leading to challenges in information sharing. To tackle this challenge, this paper develops a two‐level evolutionary game model between vehicle manufacturers and chip suppliers. It examines the effects of government rewards and penalties, trust risks, and spillover benefits from information sharing on the strategic decisions of both parties. Furthermore, the paper investigates system evolution strategies under three distinct government policies: forced, fair, and favored. The findings show that as trust risks escalate, both entities in the supply chain tend to withhold information due to concerns over losing their unique competitive advantages. However, enhancing spillover benefits through information sharing positively impacts the mitigation of trust risks. Numerical simulations demonstrate that while a government‐forced policy with increased penalties may boost information sharing in the short term, it proves less effective over the long term. Conversely, preferred rewards and subsidies under equitable policies can significantly boost one party's readiness to share information, thus enhancing cooperation between the two parties. The paper also offers countermeasures and recommendations. For instance, it suggests that upstream and downstream enterprises should not only foster mutual trust but also vigilantly track government policy directions to develop effective information‐sharing strategies. Furthermore, government industrial policies should fully consider the actual market conditions to alleviate trust risks between upstream and downstream entities and promote information sharing among enterprises.

A generalized bilevel optimization model for large-scale task scheduling in multiple agile earth observation satellites

Scheduling large-scale tasks within a multi-agile earth observation satellite system poses a formidable challenge. Given the complexity of this optimization problem, the responding solving strategy should exhibit both efficiency and effectiveness in terms of computational time and solution quality. The current strategies can broadly be categorized into all-in-one and two-stage methodologies. The latter, more conducive to real-world scenarios, owing to reducing the problem complexity and practical operational flexibility, dissects this challenge into task allocation and single-satellite scheduling. However, existing two-stage strategies still consider the objectives and constraints at the equivalent level and only adapt to specific algorithms. In this way, the pivotal issues concerning the solution complexity and strategy compatibility remain fundamentally unaddressed. To address this limitation, we proposed a generalized bilevel optimization model called Question-and-Answer model, which establishes two distinct optimization model with different contains and objectives. In this model, the upper questions are highly indispensable in the response from the lower level and constraints are considered separately at two stages. To ascertain the generalization of this framework, we conduct a comprehensive range of experiments employing various proposed strategies and algorithms in two stages. Diverse algorithms ranging from heuristic principles, and evolutionary strategies, to reinforcement learning can be seamlessly combined and integrated within this framework. The results demonstrate that the scale of scenarios does not affect the effectiveness of any amalgamated algorithms within this framework. Furthermore, the transition of upper questions according to the lower answers indeed improves the objectives but concurrently intensify the computational time.

2-D deployment of aerial base stations: A simulation model to provide voice communication

Unmanned aerial vehicles (UAVs) offer a potential alternative for providing voice services in areas where communication is disrupted due to natural disasters. These UAVs can be configured as aerial base stations (ABSs), enabling the deployment of a temporary communications network. However, communication networks based on ABSs pose several significant challenges. One of these challenges involves addressing interruptions or limitations in network coverage caused by natural disasters. In such situations, there is a high likelihood that users within the affected area may be unable to communicate due to a lack of coverage. This is a complex problem because it depends on factors, such as the mobile user locations, the characteristics of the air-to-ground channel, and geographical details of the area. In this work, we propose an optimization model to determine the placement of a set of ABSs within a limited disaster area that maximizes the probability of successful voice services (PSVSs). This optimization model integrates a network evaluation model that analyzes the wireless environment at a specific time. The network evaluation model utilizes two-ray and Rayleigh channel models, enabling the simulation of a worst-case scenario for wireless communication systems. We evaluate the proposed optimization model using the (1+1)-evolution strategy with a one-fifth success rule. We explore various parameter configurations to understand their impact on algorithm performance. This analysis helps identify the configuration of the optimization model that yields the maximum PSVSs. Simulation results indicate that by appropriately configuring the evolution strategy algorithm and comparing random ABS locations with those determined ABS locations by the evolution strategy algorithm, the PSVS can be enhanced by an average of 60%.

Invasion of the four kingdoms: the parasite journey across plant and non‐plant hosts

ABSTRACTParasites have a rich and long natural history among biological entities, and it has been suggested that parasites are one of the most significant factors in the evolution of their hosts. However, it has been emphasized less frequently how co‐evolution has undoubtedly also shaped the paths of parasites. It may seem safe to assume that specific differences among the array of potential hosts for particular parasites have restricted and diversified their evolutionary pathways and strategies for survival. Nevertheless, if one looks closely enough at host and parasite, one finds commonalities, both in terms of host defences and parasite strategies to out‐manoeuvre them. While such analyses have been the source of numerous reviews, they are generally limited to interactions between, at most, one kingdom of parasite with two kingdoms of host (e.g. similarities in animal and plant host responses against fungi). With the aim of extending this view, we herein critically evaluate the similarities and differences across all four eukaryotic host kingdoms (plants, animals, fungi, and protists) and their parasites. In doing so, we show that hosts tend to share common strategies for defence, including both physical and behavioural barriers, and highly evolved immune responses, in particular innate immunity. Parasites have, similarly, evolved convergent strategies to counter these defences, including mechanisms of active penetration, and evading the host's innate and/or adaptive immune responses. Moreover, just as hosts have evolved behaviours to avoid parasites, many parasites have adaptations to manipulate host phenotype, physiologically, reproductively, and in terms of behaviour. Many of these strategies overlap in the host and parasite, even across wide phylogenetic expanses. That said, specific differences in host physiology and immune responses often necessitate different adaptations for parasites exploiting fundamentally different hosts. Taken together, this review facilitates hypothesis‐driven investigations of parasite–host interactions that transcend the traditional kingdom‐based research fields.

Study on Household Investment Decision of Household Photovoltaic Project Promotion -- Based on Inclusive Finance Perspective

Considering the lack of research on households' investment decisions in promoting photovoltaic projects in existing studies, this study constructs a four-party evolutionary game model based on local governments, financing institutions, photovoltaic enterprises and households from the perspective of inclusive finance. In addition, according to the influence of irrational psychological factors on the decision-making outcome of each interested party in decision-making, this study innovatively introduces regret theory to improve the evolutionary game model. Then, each subject' evolutionary stabilisation strategies and system equilibrium points are analysed explicitly using the perceived benefit function and the replicated dynamic differential equation. Finally, combined with the system dynamics method, the theoretical derivation results are verified by parameter assignment, and the sensitivity of each subject to multiple influencing factors is explored. The results show that: (1) under the current market condition, the system is stabilised in the strategy combination of Case I (subsidy, conventional finance, non-promotion, loan to build), but the ideal state can be achieved through the refinement and adjustment of policies. And there is a mutual influence between the subjects. (2) The regret-elation perception psychology of each subject has a non-negligible influence on their decision-making behaviour. The combination of strategies in Case I is a conservative decision by the subjects. (3) A variety of factors influences the behavioural decisions of each subject, and the amount of rent that photovoltaic enterprises can provide is a crucial factor affecting households' investment decisions. (4) Local governments incentives and penalties play a key incentive role in promoting the active participation of financing institutions and photovoltaic enterprises in the promotion endeavour. This study provides practical guidance and policy insights for promoting the diffuse use of renewable energy and the promotion of household photovoltaic projects while enriching the applied research by combining regret theory and evolutionary game theory.

Adapting the population size in CMA-ES using nearest-better clustering method for multimodal optimization

The covariance matrix adaptation evolution strategy (CMA-ES) is one of the advanced algorithms for solving continuous single-objective optimization. In this algorithm, the population size is an important parameter which represents the number of candidate points generated in each iteration. A larger population size allows for better exploration of the search space and is typically required for handling multimodal functions, while a smaller population size facilitates quicker convergence. Adapting the population size in this algorithm raises a question about how to increase/decrease/keep it in an efficient way to solve specific problems. In this paper, we propose a novel approach for this purpose based on the information of niches observed during the evolution process. Firstly, the nearest-better clustering (NBC) technique is employed to detect the number of niches in each iteration, then the information will be used to adapt the population size accordingly. Additionally, a quasi-Newton method, which serves as a local search, can be incorporated into the algorithm in order to speed it up. We evaluate the effectiveness of our method through numerical simulations on multimodal test functions, the black-box optimization benchmarking (BBOB) noiseless testbed, and the CEC 2014 and CEC 2017 benchmark testing suites.

Enhancing power quality in solar-wind grid-connected systems through soft computing techniques

This work intends to improve estimates of solar and wind energy generation through the application of resilient backpropagation control and substantial power evolution strategy (SPES) algorithms. In comparison to particle swarm optimization and genetic algorithms, the main goal is to minimize predicting mistakes. These methods increase grid reliability by lowering total harmonic distortion (THD) and improving power quality when integrated with the IEEE-9 bus standard. In order to evaluate the hybrid system's transient and steady-state reactions, the study also highlights the importance of bolstering operation and control. A revolutionary deep learning-based approach is also suggested for predicting wind and solar hybrid energy. The power grid's efficiency and dependability in handling renewable energy sources have significantly improved, according to the results.

Soil seed bank resilience in passively restored endangered Sand Fynbos following a century of pine plantations

Societal Impact StatementEcosystems are rapidly being transformed, pushing us towards irreversible losses and even extinctions. The Kunming‐Montreal Global Biodiversity Framework aims to curb biodiversity decline. An intriguing solution lies in seed banks—where plants store seeds in the soil. Restoration efforts can revive lost ecosystems by leveraging these seed banks. In the fynbos of South Africa, this study found that it is possible to bring back ecosystems that were lost as long ago as 100 years if conditions are right. Managers can achieve best results by applying a dry season prescribed burn following removal of the driver of degradation (e.g. pine plantations or invasions).Summary Soil‐stored seed banks are a critical evolutionary strategy for plants as they can stabilize population dynamics in response to environmental fluctuations such as droughts and natural disturbances such as fires, thus improving ecosystem resilience. This study aimed to assess resilience during the first restoration cycle of endangered Sand Fynbos following pine harvesting and a prescribed burn in South Africa. We compared above ground populations and soil‐stored seed bank composition and focused on six perennial fynbos focal species in five 1 m2 plots. All six species are obligate reseeders with soil‐stored seed banks but represent a variety of important fynbos growth forms and seed types. We found that native soil‐seed bank density following over a century of plantation forestry and 10 years after the first fynbos restoration burn was comparable to, or exceeded, densities measured in other fynbos studies, especially seed banks of alien‐invaded fynbos ecosystems. The early years following pine harvesting and a restoration burn likely provide a pest‐ and predator‐free window of opportunity in which plants may re‐establish and flourish. Our results show that even for old forestry plantation areas, ecological restoration to the historic vegetation community is feasible, where there is retained seed bank resilience, through a combination of passive and active restoration methods. Ecological restoration in cases where soil seed banks persist is therefore a viable and critically important method to attain 30% conservation of terrestrial land in ecosystems with less than 30% remaining, as part of the Kunming‐Montreal Global Biodiversity Framework.

AutoLDT: a lightweight spatio-temporal decoupling transformer framework with AutoML method for time series classification

Time series classification finds widespread applications in civil, industrial, and military fields, while the classification performance of time series models has been improving with the recent development of deep learning. However, the issues of feature extraction effectiveness, model complexity, and model design uncertainty constrain the further development of time series classification. To address the above issues, we propose a Lightweight Spatio-Temporal Decoupling Transformer framework based on Automated Machine Learning technique (AutoLDT). The framework proposes a novel lightweight Transformer with fuzzy position encoding, TS-separable linear self-attention mechanism, and convolutional feedforward network, which mine the temporal and spatial features, as well as the local and global relationship of time series. Fuzzy positional encoding integrates fuzzy ideas to enhance the generalization performance of model information mining. TS-separable linear self-attention mechanism and convolutional feedforward network achieve feature extraction in a lightweight way by decoupling temporal and spatial features of time series. Notably, we adopt the Covariance Matrix Adaptation Evolution Strategy and global adaptive pruning technique to realize automated network structure design, which further improves the model training efficiency and automation, and avoids the uncertainty problem of network design. Finally, we validate the effectiveness of the proposed framework on publicly available UCR and UEA time series datasets. The experimental results show that the proposed framework not only improves the model classification performance in a lightweight way but also dramatically improves the model training efficiency.

Two-stage bidirectional coevolutionary algorithm for constrained multi-objective optimization

Objective optimization and constraint satisfaction are two primary and conflicting tasks in solving constrained multi-objective optimization problems (CMOPs). To better trade off them, this paper proposes a two-stage bidirectional coevolutionary algorithm, termed C-TBCEA, for constrained multi-objective optimization. It consists of two stages, with each concentrating on specific targets, i.e., the first stage primarily focuses on objective optimization while the second stage focuses on constraint satisfaction by employing different evolutionary strategies at each stage. Via the synergy of the two stages, a dynamic trade-off between objective optimization and constraint satisfaction can be achieved, thus overcoming the distinctive challenges that may be encountered at different stages of evolution. In addition, to take advantage of both feasible and infeasible solutions, we employ two populations, i.e., the main population that stores the non-dominated feasible solutions and the archive population that maintains the informative infeasible solutions, to prompt the bidirectional coevolution of them. To validate the effectiveness of the proposed C-TBCEA, experiments are carried out on 6 CMOP test suites and 17 real-world CMOPs. The results demonstrate that the proposed algorithm is very competitive with 9 state-of-the-art constrained multi-objective optimization evolutionary algorithms (CMOEAs).

Enhancing Phishing Detection in Semantic Web Systems Using Optimized Deep Learning Models

Phishing detection in Semantic Web systems is crucial to safeguarding users from malicious attacks. In this context, this work presents a deep learning-based phishing attack detection model using MobileBERT for feature extraction and hyperparameter optimization using covariance matrix adaptation evolution strategy (CMA-ES). The model obtained a 95% classification accuracy. Important benchmarks like accuracy, recall, and F1-score show good ability to discriminate between phishing and legitimate emails. Applying CMA-ES, which improved detection accuracy, helps to verify the model even more. MobileBERT and CMA-ES together offer Semantic Web systems a fresh, efficient method of phishing detection.

A Collaborative Auto-Diversified Optimization Scheme

We present a Collaborative Auto-Diversified Optimization Scheme (CADOS) for solving continuous and combinatorial optimization problems. CADOS aims to explore the synergy of various optimization algorithms and enhance their effectiveness and efficiency, particularly in higher-dimensional problems. It incorporates an enhanced version of the previously proposed approach Auto-Diversified Ameliorated MultiPopulation-based Ensemble Differential Evolution (AD-AMPEDE), Covariance Matrix Adaptation Evolutionary Strategy (CMA-ES), and a local search (LS) algorithm. AD-AMPEDE has demonstrated good performance in solving continuous optimization problems. However, its competitiveness wanes in higher dimensions. CADOS improves AD-AMPEDE’s detection/re-diversification processes and parameters adaptation, making it effective for higher-dimensional problem classes. To explore nearby regions during stagnation, a trust-region local search is employed. For re-diversification, CADOS utilizes both CMA-ES, known for its efficiency in complex fitness landscapes, and the Auto-Enhanced Population Diversity (AEPD) technique. We tested CADOS on the COmparing Continuous Optimizers (COCO) platform and the results demonstrated excellent performance of CADOS. In addition, to show the proposed scheme’s efficacy in tackling real-world issues, we employed it to optimize the design of water distribution networks (WDS). The results we obtained underscore the remarkable competitiveness of our strategy when compared to widely recognized existing algorithms.

Research on talent cultivation for rural revitalization based on three-party evolutionary game.

The sustainable development of Rural Revitalization Talent Training (RRTT) is a key prerequisite for realizing the rural revitalization strategy. In order to study the influence of various stakeholders on RRTT, explore its optimal development path, and clarify the key control factors, this study analyzes the behavioral decision-making of RRTT stakeholders for the first time, and constructs an evolutionary game model of the government, universities and village collectives. Through dynamic decision replication analysis and evolutionary stability analysis, the game relationship among the three stakeholders is discussed. Furthermore, through numerical simulation, the decision-making characteristics of the three parties "behaviors and the evolution trend of the stakeholders" behaviors under the current situation are evaluated, the sensitivity of the key control factors with the policy changes is analyzed, and the feasibility of its implementation is discussed. The results show that the input cost of village collectives is the main determinant of RRTT, and appropriate financial input and low incentive policies are more conducive to universities and village collectives to actively promote RRTT. The research results provide decision-making basis for the implementation of rural revitalization strategy.

Spectral-energy efficiency tradeoff of massive MIMO by a constrained large-scale multi-objective algorithm through decision transfer

To better balance the spectral efficiency (SE) and energy efficiency (EE) in the massive multiple-input multiple output system with a large number of users (MaMIMO-LU), the SE-EE tradeoff is originally constructed as a constrained large-scale multi-objective problem (CLSMOP) for the power allocation of users. To solve this CLSMOP, a constrained large-scale multi-objective evolutionary algorithm (CLSMOEA), considering the dimensionality reduction as well as the balance of objectives and constraints, is explored. The Lagrange multiplier is first used to construct a two-scale optimization model, bridging original large-scale decision space of variables and small-scale decision space of coefficients of Lagrange multiplier. The decision transfer algorithm is then designed to switch between large-scale original decision space and small-scale parametric decision space, while achieving the maximum dimensionality reduction. Finally, the two-scale evolution strategy is proposed for the alternative optimizations in the two decision spaces emphasizing objectives and constraints, respectively. In summary, the optimization in large-scale space pushes the population to unconstrained Pareto front (PF), the optimization in small-scale space helps the population cross the infeasible areas to approach constrained PF, and the GD-based reproduction of offspring further guarantees the solution convergence. Ten representative and state-of-the-art constrained multi-objective evolutionary algorithms (MOEAs) and unconstrained MOEA have been compared to the proposed CLSMOEA to demonstrate its effectiveness through comparative experiments on some well-known benchmark problems (with 1000 variables), and MaMIMO-LU (with 1024 antennas and 256, 512, and 1024 users). Experimental results show that the proposed CLSMOEA can obtain the best SE-EE tradeoff.