Evolutionary Search Research Articles

Multiobjective Differential Evolution Algorithm Balancing Multiple Stakeholders for Low-Carbon Order Scheduling in E-Waste Recycling

Order scheduling is an important part of the e-waste recycling process, which can influence the quantity and efficiency of the recycling. With the sustainable development of e-waste recycling, low-carbon order scheduling becomes a significant and challenging reverse logistics scheduling problem. However, it is difficult to obtain an effective low-carbon order schedule considering the conflicting interests of the multiple stakeholders, including enterprises, drivers, customers, and governments. To address this issue, a multi-objective order scheduling model (MOOSM) and a multi-objective differential evolution algorithm balancing multiple stakeholders (MODE-MS) are proposed in this paper. First, to embody the interests of different stakeholders, three time-dependent key variables are calculated by the road congestion and vehicle load, including the velocity, travelling time and carbon emission. Second, with the above key variables, a five-objective order scheduling model is formulated to describe the low-carbon order scheduling problem in e-waste recycling. Third, for solving the MOOSM, a multi-objective differential evolution algorithm based on an adaptive evolutionary search strategy is developed to obtain the low-carbon and stakeholders satisfied scheduling schemes. The experimental results validate the feasibility of MOOSM and the effectiveness of MODE-MS. By comparing with four state-to-art algorithms, the advantages of the proposed MODE-MS are further demonstrated in solving the low-carbon order scheduling.

Improving multi-objective evolutionary algorithms using Grammatical Evolution

Multi-objective evolutionary algorithms (MOEAs) have become an effective choice to solve multi-objective optimization problems (MOPs). However, it is well known that Pareto dominance-based MOEAs struggle in MOPs with four or more objective functions due to a lack of selection pressure in high dimensional spaces. The main choices for dealing with such problems are decomposition-based and indicator-based MOEAs. In this work, we propose the use of Grammatical Evolution (an evolutionary computation search technique) to generate functions that can improve decomposition-based and indicator-based MOEAs. Namely, we propose a methodology to generate new scalarizing functions, which are known to have a great impact in the performance of decomposition-based MOEAs and in some indicator-based MOEAs. Additionally, we propose another methodology to generate hypervolume approximations, since the hypervolume is a popular performance indicator used not only in indicator-based MOEAs but also to assess performance of MOEAs. Using our first methodology, we generate two new scalarizing functions and provide their corresponding experimental validation to show that they exhibit a competitive behavior when compared against some well-known scalarizing functions such as ASF, PBI and the Tchebycheff scalarizing function. Using our second methodology, we produce 4 different hypervolume approximations and compare their performance against the Monte Carlo method and against two other state-of-the-art hypervolume approximations. The experimental results show that our functions exhibit a good compromise in terms of quality and execution time.

Evaluating the choice of radial basis functions in multiobjective optimal control applications

Evolutionary Multi-Objective Direct Policy Search (EMODPS) is a prominent framework for designing control policies in multi-purpose environmental systems, combining direct policy search with multi-objective evolutionary algorithms (MOEAs) to identify Pareto approximate control policies. While EMODPS is effective, the choice of functions within its global approximator networks remains underexplored, despite their potential to significantly influence both solution quality and MOEA performance. This study conducts a rigorous assessment of a suite of Radial Basis Functions (RBFs) as candidates for these networks. We critically evaluate their ability to map system states to control actions, and assess their influence on Pareto efficient control policies. We apply this analysis to two contrasting case studies: the Conowingo Reservoir System, which balances competing water demands including hydropower, environmental flows, urban supply, power plant cooling, and recreation; and The Shallow Lake Problem, where a city navigates the trade-off between environmental and economic objectives when releasing anthropogenic phosphorus. Our findings reveal that the choice of RBF functions substantially impacts model outcomes. In complex scenarios like multi-objective reservoir control, this choice is critical, while in simpler contexts, such as the Shallow Lake Problem, the influence is less pronounced, though distinctive differences emerge in the characteristics of the prescribed control strategies.

Lightweight multi-objective evolutionary neural architecture search with low-cost proxy metrics

Multi-Objective Evolutionary Neural Architecture Search (MOENAS) methods employ evolutionary algorithms to approximate a set of architectures representing optimal trade-offs between network performance and complexity. Directly estimating network performance via error rates or losses incurs long runtimes due to the computationally expensive network training procedure. Instead, low-cost metrics that require no network training have been proposed as a proxy for network performance. However, these metrics might exhibit inconsistent correlations with network performance across different search spaces. The influences of training-based and training-free metrics on the effectiveness and efficiency of MOENAS are still under-explored.We introduce the Enhanced Training-Free MOENAS (E-TF-MOENAS) that employs the widely-used NSGA-II as the search algorithm and optimizes multiple training-free performance metrics as separate objectives. Experiments on NAS-Bench-101 and NAS-Bench-201 show that E-TF-MOENAS outperforms training-free methods that use a single training-free performance metric and could obtain comparable results to training-based methods but with approximately 30 times less computation cost. E-TF-MOENAS obtains architectures in NAS-Bench-201 with state-of-the-art mean accuracies of 94.37%, 73.50%, and 46.62% for CIFAR-10, CIFAR-100, and ImageNet16-120, respectively, within less than 3 GPU hours. It is beneficial to utilize multiple training-free proxy metrics simultaneously and E-TF-MOENAS provides a convenient framework for building such an efficient NAS approach. The source code can be found at https://github.com/ELO-Lab/E-TF-MOENAS.

Using machine learning to mine mental health diagnostic groups from emergency department presentations before and during the COVID-19 pandemic

PurposeThe COVID-19 pandemic had a profound negative effect on mental health worldwide. The hospital emergency department plays a pivotal role in responding to mental health crises. Understanding data trends relating to hospital emergency department usage is beneficial for service planning, particularly around preparing for future pandemics. Machine learning has been used to mine large volumes of unstructured data to extract meaningful data in relation to mental health presentations. This study aims to analyse trends in five mental health-related presentations to an emergency department before and during, the COVID-19 pandemic.MethodsData from 690,514 presentations to two Australian, public hospital emergency departments between April 2019 to February 2022 were assessed. A machine learning-based framework, Mining Emergency Department Records, Evolutionary Algorithm Data Search (MEDREADS), was used to identify suicidality, psychosis, mania, eating disorder, and substance use.ResultsWhile the mental health-related presentations to the emergency department increased during the COVID-19 pandemic compared to pre-pandemic levels, the proportion of mental health presentations relative to the total emergency department presentations decreased. Several troughs in presentation frequency were identified across the pandemic period, which occurred consistently during the public health lockdown and restriction periods.ConclusionThis study implemented novel machine learning techniques to analyse mental health presentations to an emergency department during the COVID-19 pandemic. Results inform understanding of the use of emergency mental health services during the pandemic, and highlight opportunities to further investigate patterns in presentation.

Surrogate-assisted level-based learning evolutionary search for geothermal heat extraction optimization

An enhanced geothermal system is essential to provide sustainable and long-term geothermal energy supplies and reduce carbon emissions. Optimal well-control scheme for effective heat extraction and improved heat sweep efficiency plays a significant role in geothermal development. However, the optimization performance of most existing optimization algorithms deteriorates as dimension increases. To solve this issue, a novel surrogate-assisted level-based learning evolutionary search algorithm (SLLES) is proposed for heat extraction optimization of enhanced geothermal system. SLLES consists of classifier-assisted level-based learning pre-screen part and local evolutionary search part. The cooperation of the two parts has realized the balance between the exploration and exploitation during the optimization process. After iteratively sampling from the design space, the robustness and effectiveness of the algorithm are proven to be improved significantly. To the best of our knowledge, the proposed algorithm holds state-of-the-art simulation-involved optimization framework. Comparative experiments have been conducted on benchmark functions, a two-dimensional fractured reservoir and a three-dimensional enhanced geothermal system. The proposed algorithm outperforms other five state-of-the-art surrogate-assisted algorithms on all selected benchmark functions. The results on the two heat extraction cases also demonstrate that SLLES can achieve superior optimization performance compared with traditional evolutionary algorithm and other surrogate-assisted algorithms. This work lays a solid basis for efficient geothermal extraction of enhanced geothermal system and sheds light on the model management strategies of data-driven optimization in the areas of energy exploitation.

Strength and manufacturability enhancement of a composite automotive component via an integrated finite element/artificial neural network multi-objective optimization approach

This study addresses the enhancement of an injection-molded fiber-reinforced plastic / metal hybrid automotive structure and its plastic injection molding process through the integration of the finite element method, artificial intelligence and evolutionary search methods. Experiments are conducted to validate the finite element models. The orthogonal array and Latin hypercube methods are employed to generate a database via finite element analysis. The database is then used to train artificial neural networks that accurately evaluate component distortion, manufacturing time, and structural strength. A genetic optimization algorithm is applied to identify optimal process parameters. The procedure was demonstrated to simultaneously reduce product warpage and manufacturing time by 10 and 62 %, respectively, when compared with the reference manufacturing process while strength is kept above the required levels with a reduced number of required data points. A more in-depth investigation into the causes of strength variation and deformation is also provided. The results contribute to the advance of robust composite automotive structures with superior quality, manufactured through efficient processes.

Learning Regularity for Evolutionary Multiobjective Search: A Generative Model-Based Approach

Learning Regularity for Evolutionary Multiobjective Search: A Generative Model-Based Approach

Gibbon: An Efficient Co-Exploration Framework of NN Model and Processing-In-Memory Architecture

The memristor-based Processing-In-Memory (PIM) architectures have been proven to be a potential architecture to store enormous parameters and execute the complicated computations of Deep Neural Networks (DNNs) efficiently. Existing PIM studies focus on designing high energy-efficient hardware architecture and algorithm-hardware co-optimization for better performance. However, the impacts of the algorithms and hardware architectures on the performance intersect with each other. Only optimizing the algorithms or the hardware architectures can not realize the optimal design. Therefore, the co-exploration of NN models and PIM architecture is necessary. However, for one thing, the co-exploration space size of NN models and PIM architectures is extremely huge, and is challenging to search. For another, during the co-exploration process, time-consuming PIM simulators are needed to evaluate various design candidates and pose a heavy time burden. To tackle these problems, we propose an efficient co-exploration framework of NN models and PIM architectures, named . In, the co-exploration space is carefully designed to adapt both NN models and PIM architectures. Besides, in order to improve search efficiency, we propose an evolutionary search algorithm with adaptive parameter priority (ESAPP). In addition, introduces a multi-level joint simulator to alleviate the problem of time-consuming evaluation. The experimental results show that the proposed co-exploration framework can find better NN models and PIM architectures than existing studies in only six GPU hours (9.8 48.2× speedup). At the same time, can improve the accuracy of co-design results by 15.3% and reduce the energy-delay-product (EDP) by 5.96× compared with existing work.

Scatter search for capacitated minimum spanning tree

We develop a scatter search algorithm for the classical capacitated minimum spanning tree (CMST) problem. We address both the homogeneous and the more difficult heterogeneous variant of the problem. The CMST is central in a wide range of network design applications in industrial engineering, routing and logistics, and communication networks. Since the problem is NP-Complete, heuristic solution methods are often required to find high-quality solutions within practical time limits. Research efforts have been focused on complex algorithm designs combining advanced neighborhood search techniques with exact solution methods or hybrids of genetic algorithms with various alternative search techniques and likewise supplemented with optimization procedures. In this study we go back to the roots and offer an alternative algorithm that competes with the best algorithms in the literature and both avoids complicated artifacts and is simpler to implement. Key features of the proposed algorithm include an oscillation neighborhood search method characterized by two classes of neighborhoods and an evolutionary search procedure based on a scatter search framework.

The Detection of Colorectal Cancer through Machine Learning-Based Breath Sensor Analysis.

Colorectal cancer (CRC) is the third most common malignancy and the second most common cause of cancer-related deaths worldwide. While CRC screening is already part of organized programs in many countries, there remains a need for improved screening tools. In recent years, a potential approach for cancer diagnosis has emerged via the analysis of volatile organic compounds (VOCs) using sensor technologies. The main goal of this study was to demonstrate and evaluate the diagnostic potential of a table-top breath analyzer for detecting CRC. Breath sampling was conducted and CRC vs. non-cancer groups (105 patients with CRC, 186 non-cancer subjects) were included in analysis. The obtained data were analyzed using supervised machine learning methods (i.e., Random Forest, C4.5, Artificial Neural Network, and Naïve Bayes). Superior accuracy was achieved using Random Forest and Evolutionary Search for Features (79.3%, sensitivity 53.3%, specificity 93.0%, AUC ROC 0.734), and Artificial Neural Networks and Greedy Search for Features (78.2%, sensitivity 43.3%, specificity 96.5%, AUC ROC 0.735). Our results confirm the potential of the developed breath analyzer as a promising tool for identifying and categorizing CRC within a point-of-care clinical context. The combination of MOX sensors provided promising results in distinguishing healthy vs. diseased breath samples. Its capacity for rapid, non-invasive, and targeted CRC detection suggests encouraging prospects for future clinical screening applications.

Theta-mechanism based cluster search algorithm for global constrained optimization

This study concerns constructing an evolutionary search system to solve the global constrained optimization problems. Firstly, we proposed a hybrid constraint-handling method, called theta-mechanism, which blends two types of constraint-handling functions and alternates use of them in the searching process to balance two competing objectives: seeking as much as possible feasible regions and quickly converging to the optimum point in the found feasible regions. Secondly, to enable the search system to cooperate well with the theta-mechanism, we designed the cluster search algorithm (CSA) and developed the search reachability analysis (SRA) method. Based on SRA, we evaluated the characteristics of several typical search operators in order to assemble them into different operator combinations in CSA to maximize its performance, which enables CSA with theta-mechanism to accomplish the two inconsistent search objectives effectively. We tested the proposed method on 18 benchmark functions from IEEE CEC2010 and 32 real-world constrained optimization problems collected in IEEE CEC2020. Our results show the CSA with theta-mechanism is more competitive than the existing state-of-the-art approaches.

Structurally Constrained Evolutionary Algorithm for the Discovery and Design of Metastable Phases.

Metastable materials are abundant in nature and technology, showcasing remarkable properties that inspire innovative materials design. However, traditional crystal structure prediction methods, which rely solely on energetic factors to determine a structure's fitness, are not suitable for predicting the vast number of potentially synthesizable phases that represent a local minimum corresponding to a state in thermodynamic equilibrium. Here, we present a new approach for the prediction of metastable phases with specific structural features and interface this method with the XtalOpt evolutionary algorithm. Our method relies on structural features that include the local crystalline order (e.g, the coordination number or chemical environment), and symmetry (e.g, Bravais lattice and space group) to filter the breeding pool of an evolutionary crystal structure search. The effectiveness of this approach is benchmarked on three known metastable systems: XeN8, with a two-dimensional polymeric nitrogen sublattice, brookite TiO2, and a high pressure BaH4 phase, which was recently characterized. Additionally, a newly predicted metastable melaminate salt, P1̅ WC3N6, was found to possess an energy that is lower than that of two phases proposed in a recent computational study. The method presented here could help in identifying the structures of compounds that have already been synthesized, and in developing new synthesis targets with desired properties.

Development of hybrid surrogate model structures for design and optimization of CO2 capture processes: Part I. Vacuum pressure swing adsorption in a confined space

Design and optimization of CO2 capture processes have become a tremendously active area of research particularly in the past decade. In this context, development of intelligent techniques on the basis of first-principles models coupled with data-driven algorithms for such purpose looks very promising. In a series of works, we intend to present mechanism approaches in order to develop applicable structures for design and optimization of the CO2 capture processes of interest via hybrid models. A systematic method is presented for optimizing a process for capturing CO2 from a confined space through a Vacuum Pressure Swing Adsorption (VPSA) operation using a Hybrid Surrogate Model (HSM) and Non-dominated Sorting Genetic Algorithm (NSGA-III). The surrogate model is structured based on the VPSA process model offered in the Aspen Adsorption® environment and an artificial intelligence (AI) data-driven algorithm. The developed HSM is then used to predict the key process outputs including CO2 purity, air recovery ratio and energy consumption rate. Accordingly, the optimized parameters are re-substituted into the VPSA process simulator for further data processing. It is demonstrated that the proposed model architecture provides considerable computational efficiency for the process optimization with only 48 h to complete the corresponding evolutionary search, while the optimization time by the conventional NSGA-direct method is close to 1129 h. The optimization results also show that the CO2 purity changes from 1000 ppm to 399 ppm, the air recovery ratio remains at 93 %, and the energy consumption per unit product (ECP) decreases by 38.5 % to 99.7 kJ·Nm−3 air after an optimized air purification operation. The idea of chemical mechanism and industrial data twin modeling in this study holds substantial importance for the development of digital chemical twin systems and the process optimization of intelligent factory.

A Boosted Evolutionary Neural Architecture Search for Timeseries Forecasting with Application to South African COVID-19 Cases

In recent years, there has been an increase in studies on time-series forecasting for the future occurrence of disease incidents. Improvements in deep learning approaches offer techniques for modelling long-term temporal relationships. Nonetheless, this design practice is rigorously painstaking, prone to errors, and requires human expertise. The advent of feature enrichment with automatic architecture search typically optimises the discovery of new neural architectures applicable in domains such as time-series modelling. The main methodological contribution of this study is an approach for time-series forecasting using feature-enriched filters and an evolutionary neural architecture search with sequence-to-sequence gated recurrent units (GRU-Seq2Seq). This is applied to the prediction of daily cases of coronavirus disease in South Africa. The highly pathogenic coronavirus pandemic incident data was modelled with filters, optimised hyper-parameter search trials and an evolutional neural algorithm. The proposed model was benchmarked against ARIMA and SARIMA. The model predicted trends for 30, 60 and 90-day horizons and evaluated them for 7, 14 and 31 days. Simulation results demonstrate that observed daily case counts with added filters and evolutionary search optimisation for forecasting improve performance accuracy. Generally, the proposed bFilter+GRU-Seq2Seq with optimal search configuration outperformed ARIMA and SARIMA with lower error scores and higher performance metrics, with an R2 score of 7.48E-01 for a 30-day forecast horizon.

Evolving Interactive Narrative Worlds

An interactive narrative is bound by the context of the world where its story takes place. However, most work in interactive narrative generation takes its story world design and mechanics as given, which abdicates a large part of story generation to an external world designer. In this paper, we close the story world design gap with an evolutionary search framework for generating interactive narrative worlds and mechanics. Our framework finds story world designs that accommodate multiple distinct player roles. We evaluate our system with an action agreement ratio analysis that shows worlds generated by our framework provide a greater number of in-role action opportunities compared to story worlds randomly sampled from the generative space.

Seasonal forecast-informed reservoir operation. Potential benefits for a water-stressed Mediterranean basin

The increased seasonal demand for water puts pressure on Mediterranean water resources, which are often exploited in a non-renewable way. Besides, climate change can alter hydroclimatic patterns and exaggerate freshwater stress. Flexible operation of existing water reservoirs is one of the most cost-effective ways to mitigate water-related stress by storing water when it is abundant and releasing it when droughts persist. In this context, hydroclimatic forecasts can be central in properly informing reservoir operation. Nevertheless, the link between forecast skill and forecast value is neither easily predictable nor necessarily positive. Each system requires specific forecasts according to its characteristics, and the skill of existing forecast systems does not necessarily translate into a significant gain in system performance. In this work, we develop downscaled seasonal forecasts of reservoir inflow for the Faneromeni irrigation dam on Crete island. We then quantify the value of these seasonal forecasts in informing the reservoir operations. While the current operation of this reservoir is based on the available storage at the beginning of the irrigation season, we investigate alternatives by using the Evolutionary Multi Objectives Direct Policy Search method, which allows the design of flexible rules to cope with the variability of the hydrologic conditions as well as to include forecast information for conditioning operational decisions. Under historical climate conditions, results demonstrate a notable enhancement in performance solely by implementing more flexible operating policies. Incorporating perfect forecasts results in an additional improvement of 4% on average throughout the period from 1993 to 2019. However, when using actual forecasts, this gain diminishes to 1%. These outcomes support the exploration of potential trade-off solutions that effectively balance the competing demands within the region.

Solving an energy resource management problem with a novel multi-objective evolutionary reinforcement learning method

Microgrids have become popular candidates for integrating diverse energy sources into the power grid as means of reducing fossil fuel usage. Energy Resource Management (ERM) is a type of Unit Commitment problem, where a player operates a microgrid with diverse renewable generators integrated with an external supplier. Calculating the economic dispatch of each committed unit on a planning horizon is an NP-hard problem, and therefore, finding an exact solution is difficult. This paper presents a multi-objective solution to the ERM problem from the perspective of battery operation and external supplier dispatch. First, a novel multi-objective decision problem modeling is proposed that considers three objectives: cost, greenhouse gas emissions, and battery degradation. This framework involves a learning agent that controls the depth of discharge of a Lithium-Ion battery. To address the proposed problem, a new multi-objective algorithm called Multi-Objective Evolutionary Policy Search (MEPS) is introduced. The proposed algorithm uses NeuroEvolution of Augmenting Topologies structure to evolve artificial neural networks for estimating action-preference values considering multi-objective rewards. The MEPS performance is evaluated on both standard and newly-proposed benchmark problems, using the hypervolume as the evaluation metric. When compared to standard deep reinforcement learning, results showed that MEPS provides cost-effective, environmentally friendly, and efficient energy storage management solutions. Furthermore, MEPS effectively solves the proposed ERM problem by finding neural networks with a small number of nodes and connections, which are suitable for use in embedded control systems. Overall, MEPS proved to be a promising multi-objective approach in the transition to clean energy resources.

Efficient multi-objective evolutionary neural architecture search for U-Nets with diamond atrous convolution and Transformer for medical image segmentation

Deep encoder–decoder neural networks like U-Nets have made significant contributions to the development of computer vision applications such as image segmentation. Neural architecture search (NAS) has the potential to further automatically adjust the architectures of U-Nets for various medical image segmentation tasks. Most of the NAS techniques focus on optimizing segmentation accuracies of network architectures. In real-world medical image segmentation scenarios, two main challenges are poor medical image quality and diverse deployment devices with different computing capabilities. A large architecture designed only for the high segmentation accuracy is difficult to run on various deployment devices. To address these challenges, this paper proposes a multi-objective evolutionary neural architecture search method (CTU-NAS) for U-Nets with diamond atrous convolution and Transformer for medical image segmentation. A hybrid U-Net architecture (CTU-Net) with diamond atrous convolution and Transformer modules is designed as the supernet of CTU-NAS. Then a channel search strategy based on sorting and selection is applied to speed up the search for subnets by precisely selecting and training the most important channels more times. In addition, CTU-NAS employs a dual acceleration mechanism based on weight sharing and surrogate model to lower the cost of evaluations of subnets. CTU-NAS applies a multi-objective evolutionary algorithm to balance between the segmentation accuracy and the number of parameters. Experimental results on two medical segmentation datasets show that CTU-NAS is capable of quickly generating a group of excellent network architectures with different sizes and their performances also outperform or come close to those of the manually designed networks.

Evolutionary neural architecture search combining multi-branch ConvNet and improved transformer

Deep convolutional neural networks (CNNs) have achieved promising performance in the field of deep learning, but the manual design turns out to be very difficult due to the increasingly complex topologies of CNNs. Recently, neural architecture search (NAS) methods have been proposed to automatically design network architectures, which are superior to handcrafted counterparts. Unfortunately, most current NAS methods suffer from either highly computational complexity of generated architectures or limitations in the flexibility of architecture design. To address above issues, this article proposes an evolutionary neural architecture search (ENAS) method based on improved Transformer and multi-branch ConvNet. The multi-branch block enriches the feature space and enhances the representational capacity of a network by combining paths with different complexities. Since convolution is inherently a local operation, a simple yet powerful “batch-free normalization Transformer Block” (BFNTBlock) is proposed to leverage both local information and long-range feature dependencies. In particular, the design of batch-free normalization (BFN) and batch normalization (BN) mixed in the BFNTBlock blocks the accumulation of estimation shift ascribe to the stack of BN, which has favorable effects for performance improvement. The proposed method achieves remarkable accuracies, 97.24 % and 80.06 % on CIFAR10 and CIFAR100, respectively, with high computational efficiency, i.e. only 1.46 and 1.53 GPU days. To validate the universality of our method in application scenarios, the proposed algorithm is verified on two real-world applications, including the GTSRB and NEU-CLS dataset, and achieves a better performance than common methods.