Multi-objective Selection Research Articles

PyBrOpS: a Python package for breeding program simulation and optimization for multi-objective breeding.

Plant breeding is a complex endeavor that is almost always multi-objective in nature. In recent years, stochastic breeding simulations have been used by breeders to assess the merits of alternative breeding strategies and assist in decision-making. In addition to simulations, visualization of a Pareto frontier for multiple competing breeding objectives can assist breeders in decision-making. This paper introduces Python Breeding Optimizer and Simulator (PyBrOpS), a Python package capable of performing multi-objective optimization of breeding objectives and stochastic simulations of breeding pipelines. PyBrOpS is unique among other simulation platforms in that it can perform multi-objective optimizations and incorporate these results into breeding simulations. PyBrOpS is built to be highly modular and has a script-based philosophy, making it highly extensible and customizable. In this paper, we describe some of the main features of PyBrOpS and demonstrate its ability to map Pareto frontiers for breeding possibilities and perform multi-objective selection in a simulated breeding pipeline.

A compass-based hyper-heuristic for multi-objective optimization problems

Multi-objective selection hyper-heuristics have attracted more attention of researchers because of their cross-domain ability. However, for multi-objective optimization problems (MOPs), obtaining a manageable number of solutions that are well distributed and converged in the objective space is still a challenge, especially when solving high-dimensional MOPs. In order to solve this problem, this paper proposes a compass-based hyper-heuristics(COHH), which is a general iterative framework that learns and selects from a set of meta-heuristics or components (named low-level heuristics, LLHs). The selected LLH is applied to solve the given MOP at the current iteration. In order to learn the potential of LLHs, the impact of the diversity of the current solution set on the final performance is studied. Then a new compass-based indicator is defined to evaluate the current solution sets. The learning strategy with new indicator can bias to diversity by adjusting the angle of a reference vector. After learning, the adaptive two-stage selection strategy triggered by the quality of the current solution set is used to choose LLH. Experiments are conducted on DTLZ, MaOP, WFG, and MaF test suites, as well as several real-world constrained test problems. Experimental results show that COHH is competitive in performance and cross-domain capability when compared with popular meta-heuristics and hyper-heuristics.

IMF 2 O 2 : A Fully Connected Sensor Deployment Algorithm for Underwater Sensor Networks

To address the problems of node deployment schemes in existing underwater sensor networks that lack consideration of network connectivity and high deployment costs, this article constructs an optimization model that maximizes network coverage and minimizes deployment costs while ensuring full connectivity. For the NP-hard property of this optimization model, an improved moth flame optimization node deployment algorithm based on fuzzy operators (IMF 2 O 2 ) is proposed. First, comprehensively considering the two performance metrics of network coverage and network connectivity, a multi-objective selection mechanism based on fuzzy operators is proposed to improve network coverage while ensuring full connectivity. Second, a fixed number of nodes are used to monitor the target event points, transforming the node deployment of sensors into an optimal problem and proposing an improved moth flame optimization algorithm to solve this problem. Finally, the two metrics of coverage and deployment cost are measured and the fuzzy operator is used to select the optimal number of nodes to be deployed. Numerical results showed that the proposed algorithm improved network coverage rate by 10%, 22%, and 25%, and improved network connectivity rate by 12%, 20%, and 8% as compared to PSSD, RAWS, and VODA, respectively, while ensuring full connectivity.

An insight on optimization of FSP process parameters for the preparation of AA5083/(SiC-Gr) hybrid surface composites using the response surface methodology

Aluminum alloys are known for their extensive use in aerospace, automobile, marine, etc., industries due to their excellent inherent properties. Recent studies have developed different methods to modify the surface properties of aluminum by producing surface composites, such as the friction stir processing (FSP) method. The current study made an effort to develop a new hybrid surface composite of AA5083/(SiC-Gr) using the FSP method. For FSP process optimization, the response surface methodology (RSM) has been used. For creating the mathematical model using RSM, various input process parameters of the FSP are selected to predict the output characteristics of the prepared hybrid composite. A Box–Behnken design was used for the process with four factors, each factor was used with three levels, and the RSM was utilized to form a regression model to predict the responses. The ANOVA analysis suggests that NoP (number of passes): 3 and RV (reinforcement volume): 75:25 (SiC: Gr) ratio are the significant parameters of the study with a p-value less than .05. The novelty of this study lies in the development of a new hybrid surface composite of AA5083/(SiC-Gr) using the friction stir processing (FSP) method, with optimization achieved through the response surface methodology (RSM) and multi-objective selection criteria, resulting in predicted outcomes within a range of ±10% of the experimental observations.

Artificial selection methods from evolutionary computing show promise for directed evolution of microbes.

Directed microbial evolution harnesses evolutionary processes in the laboratory to construct microorganisms with enhanced or novel functional traits. Attempting to direct evolutionary processes for applied goals is fundamental to evolutionary computation, which harnesses the principles of Darwinian evolution as a general-purpose search engine for solutions to challenging computational problems. Despite their overlapping approaches, artificial selection methods from evolutionary computing are not commonly applied to living systems in the laboratory. In this work, we ask whether parent selection algorithms-procedures for choosing promising progenitors-from evolutionary computation might be useful for directing the evolution of microbial populations when selecting for multiple functional traits. To do so, we introduce an agent-based model of directed microbial evolution, which we used to evaluate how well three selection algorithms from evolutionary computing (tournament selection, lexicase selection, and non-dominated elite selection) performed relative to methods commonly used in the laboratory (elite and top 10% selection). We found that multiobjective selection techniques from evolutionary computing (lexicase and non-dominated elite) generally outperformed the commonly used directed evolution approaches when selecting for multiple traits of interest. Our results motivate ongoing work transferring these multiobjective selection procedures into the laboratory and a continued evaluation of more sophisticated artificial selection methods.

On twin screw extrusion parametric optimisation using hybrid approach of ANOVA and TOPSIS for 3D printing applications

ABSTRACT In the last decade, several studies have been performed on the preparation of smart composite matrix materials by twin-screw extrusion (TSE) followed by its process parametric optimisation based on analysis of variance (ANOVA) for 3D printing applications. But, hitherto little has been reported on TSE process optimisation using a hybrid approach, that is, ANOVA in the first stage followed by a technique for order of preference by similarity to ideal solution (TOPSIS) in the second stage for multi-objective selection. In this study, an effort has been made to develop a feedstock filament of polylactic acid (PLA) based composite matrix (comprising polyvinyl chloride (PVC) – wood dust (WD), and Fe3O4 as reinforcement) with TSE. The result of the study suggests that the proposed hybrid model of analysis is more useful than the single approach (in which the user may first use the ANOVA technique for optimisation followed by TOPSIS for locating the best and worst parametric conditions). Finally, the correlation matrix has been prepared for the mechanical and magnetic properties of the PLA-PVC-WD- Fe3O4 composite matrix.

Multi-objective uncertain project selection considering synergy

Multi-objective uncertain project selection considering synergy

Research on online-offline information resource joint regulation method of cross-border e-commerce model based on genetic algorithm

In order to overcome the problems of large adjustment error, low utilisation rate of information resources and low execution efficiency caused by traditional methods without considering task execution priority, this paper proposes a joint adjustment method of online and offline information resources based on genetic algorithm in cross-border e-commerce mode. Combined with grid technology, this method sets up an online-offline joint regulation model of information resources through multi-objective planning, and then uses multi-objective selection method to form information subgroups, and obtains new genetic population through crossover and mutation calculation, so as to obtain the optimal scheme of information resources joint regulation. The experimental results show that the maximum implementation efficiency of this method can reach 98.1%, the utilisation rate of information resources is always above 95%, and the adjustment error is always below 6%, which proves that this method is effective.

Deep Learning Test Optimization Method Using Multi-objective Optimization

PDF HTML XML Export Cite reminder Deep Learning Test Optimization Method Using Multi-objective Optimization DOI: 10.21655/ijsi.1673-7288.00282 Author: Affiliation: Clc Number: Fund Project: Article | Figures | Metrics | Reference | Related | Cited by | Materials | Comments Abstract:With the rapid development of deep learning technology, research on its quality assurance is raising more attention. Meanwhile, it is no longer difficult to collect test data owing to the mature sensor technology, but it costs a lot to label the collected data. To reduce the cost of labeling, the existing studies attempt to select a test subset from the original test set. The test subset, however, only ensures that the overall accuracy (the accuracy of the target deep learning model on all test inputs of the test set) of the test subset is similar to that of the original test set; it cannot maintain other test properties similar to those of the original test set. For example, it cannot fully cover all kinds of test input in the original test set. This study proposes a method based on multi-objective optimization called Deep Multi-Objective Selection (DMOS). It firstly analyzes the data distribution of the original test set by Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN). Then, it designs multiple optimization objectives given the characteristics of the clustering results and then carries out multi-objective optimization to find out the appropriate selection solution. Massive experiments are carried out on eight pairs of classic deep learning test sets and models. The results reveal that the best test subset selected by the DMOS method (the test subset corresponding to the Pareto optimal solution with the best performance) can not only cover more test input categories in the original test set but also estimate the accuracy of each test input category extremely close to that of the original test set. Meanwhile, it can also ensure that the overall accuracy and test adequacy are close to those of the original test set: the average error of the overall accuracy estimation is only 1.081%, which is 0.845% lower than that of Practical ACcuracy Estimation (PACE), an improvement of 43.87%. The average error of the accuracy estimation of each test input category is only 5.547%, which is 2.926% less than that of PACE, an improvement of 34.53%. The average estimation error of the five test adequacy measures is only 8.739%, which is 7.328% lower than that of PACE, an improvement of 45.61%. Reference Related Cited by

Hybrid Data Mining Forecasting System Based on Multi-Objective Optimization and Selection Model for Air pollutants

The air quality index (AQI) indicates the short-term air quality situation and changing trend of the city, which includes six air pollutants: PM2.5, PM10, CO, NO2, SO2 and O3. Due to the diversity of pollutants and the fluctuation of single pollutant time series, it is a challenging task to find out the main pollutants and establish an accurate forecasting system in a city. Previous studies primarily focused on enhancing either forecasting accuracy or stability and failed to analyze different air pollutants at length, leading to unsatisfactory results. In this study, a model selection forecasting system is proposed that consists of data mining, data analysis, model selection, and multi-objective optimized modules and effectively solves the problems of air pollutants monitoring. The proposed system employed fuzzy C-means cluster algorithm to analyze 13 original AQI series, and fuzzy comprehensive evaluation is used to find out the main air pollutants in each city. And then multiple artificial neural networks are used to forecast the main air pollutants for each category and find the optimal models. Finally, the modified multi-objective optimization algorithm is used to optimize the parameters of optimal models and model selection to obtain final forecasting values from optimal hybrid models. The experiment results of datasets from 13 cities in the Beijing–Tianjin–Hebei Urban Agglomeration demonstrated that the proposed system can simultaneously obtain efficient and reliable data for air quality monitoring.

Design and Year-Long Performance Evaluation of a Pump as Turbine (PAT) Pico-Hydropower Energy Recovery Device in a Water Network

Despite the existence of a vast scientific literature on the subject of hydraulic energy recovery from water pipelines by means of micro- and pico-hydropower using pumps as turbines (PAT), such technology has found a very limited application thus far in practice. In fact, the selection of a pump as turbine for a specific site is a matter of nontrivial trade-offs between a variety of technical and economic parameters and this aspect has not been sufficiently captured and integrated into a reliable, proven, and practical selection methodology available to designers and practitioners. Hence, a multi-objective PAT selection software was developed and utilized to design a 3 kW energy recovery installation at a rural Irish water network. The performance of the scheme was monitored over 13 consecutive months, resulting in the generation of nearly 16,000 kWh of electricity and validating the output from the PAT selection software.

The evaluation of renewable energy resources in Turkey by integer multi-objective selection problem with interval coefficient

This paper aims to provide a scientific approach that indicates the need to focus on renewable energy potential to meet energy needs in Turkey. Turkey began to take advantage of renewable energy technologies a few years ago. Accordingly, the issue of determining the best renewable resources for the country has been brought to the agenda in recent years. The issue was considered to be a limited multi-objective optimization problem allowing us to achieve a reliable result. However, the parameter of each resource was considered as a range value, rather than traditionally expressed as an exact value, and a multi-objective decision problem was developed with an interval coefficient. This method allows us to obtain more accurate and reliable results without the need to resort to normalization methods used to eliminate unit differences. According to the results of this study, the most convenient alternatives for Turkey are hydro, wind, and solar power. The findings also support decision policies aimed at reaching targets for the electricity sector in 2023, as put forth by the Ministry of Energy and Natural Resources (MENR).

MultiETSC: automated machine learning for early time series classification

Early time series classification (EarlyTSC) involves the prediction of a class label based on partial observation of a given time series. Most EarlyTSC algorithms consider the trade-off between accuracy and earliness as two competing objectives, using a single dedicated hyperparameter. To obtain insights into this trade-off requires finding a set of non-dominated (Pareto efficient) classifiers. So far, this has been approached through manual hyperparameter tuning. Since the trade-off hyperparameters only provide indirect control over the earliness-accuracy trade-off, manual tuning is tedious and tends to result in many sub-optimal hyperparameter settings. This complicates the search for optimal hyperparameter settings and forms a hurdle for the application of EarlyTSC to real-world problems. To address these issues, we propose an automated approach to hyperparameter tuning and algorithm selection for EarlyTSC, building on developments in the fast-moving research area known as automated machine learning (AutoML). To deal with the challenging task of optimising two conflicting objectives in early time series classification, we propose MultiETSC, a system for multi-objective algorithm selection and hyperparameter optimisation (MO-CASH) for EarlyTSC. MultiETSC can potentially leverage any existing or future EarlyTSC algorithm and produces a set of Pareto optimal algorithm configurations from which a user can choose a posteriori. As an additional benefit, our proposed framework can incorporate and leverage time-series classification algorithms not originally designed for EarlyTSC for improving performance on EarlyTSC; we demonstrate this property using a newly defined, “naïve” fixed-time algorithm. In an extensive empirical evaluation of our new approach on a benchmark of 115 data sets, we show that MultiETSC performs substantially better than baseline methods, ranking highest (avg. rank 1.98) compared to conceptually simpler single-algorithm (2.98) and single-objective alternatives (4.36).

Bankruptcy prediction on the base of the unbalanced data using multi-objective selection of classifiers

The goal of the paper is to develop a new algorithm for predicting whether the company will go bankrupt on the base of unbalanced data. To do it, we propose to consider the classification as a multi-objective optimization problem and construct a prediction model as an ensemble while minimizing the parameters FPR (False Positive Rate) and FNR (False Negative Rate) at the same time. To create the ensemble, the proposed algorithm of a Multi-Objective Classifier Selection (MOCS) selects only classifiers that belong to the Pareto-optimal set in FPR/FNR space; that is, there is no dominance between them, and they satisfy some additional conditions. In the general case, MOCS is determined by three parameters: two threshold values that limit false rates (FNR and FPR), and the crowding distance, which defines the uniqueness of the classifier's results. We tested the proposed algorithm on data collected from 2457 Russian companies, 456 of which went bankrupt, and 5910 Polish companies, 410 of which received bankruptcy status. Datasets contain features such as financial ratios and business environment factors. In the testing, we used more than 70 combinations of under-sampling, over-sampling, and no sampling methods with static and dynamic classification models. Final ensembles include seven classifiers for the Russian dataset and four classifiers for the Polish dataset combined by soft voting rule. In both cases, the proposed algorithm produces a significant improvement of prediction results as in terms of standard metrics (geometric mean, the area under the ROC curve) and in the visual representation in the FNR/FPR space, namely in the shift from a Pareto-optimal set of classifiers.

An evolutionary parallel multiobjective feature selection framework

Feature selection has become an indispensable preprocessing step in data mining problems as high amount of data become prevalent with the advances in technology. The objective of feature selection is twofold: reducing data amount and improving learning performance. In this study, we leverage the multi-core nature of a regular PC to build a robust framework for feature selection. This framework executes the feature selection algorithm on four processors, in parallel. As per the No Free Lunch Theorem, we facilitate 40 different execution settings for the processors by employing two multiobjective selection algorithms, four initial population generation methods, and five machine learning techniques. Besides, we introduce six setting selection schemes to decide the most fruitful setting for each processor. We carry out extensive experiments on 11 UCI benchmark datasets and analyze the results with statistical tests. Finally, we compare our proposed method with state-of-the-art studies and record remarkable improvement in terms of maximum accuracy.

Structural Optimization of Frame of the Multi-Rotor Unmanned Aerial Vehicle through Computational Structural Analysis

Because of the flexible working nature, the Multi-Rotor Unmanned Aerial Vehicle (MR-UAV) has been implemented in all kinds of real-time complicated problems. Due to this huge implementation, flawless UAVs have been required and used everywhere. Even though flawless UAVs have emerged, some of the domains in UAVs need to be analysed thoroughly to evolve UAVs without life affecting factors. This work deals with such kind of domain, which is a multi-objective investigation on MR-UAV’s airframe. In which, five different prime MR-UAV’s airframes are constructed, modelled, and computationally solved for three major composite materials and one conventional material. The important selection factors involved in this multi-objective optimization are low-stress induction, less deformation, and high lifetime. The short-listed five platforms of this structural optimization are the “X” frame, “I” frame, “K” frame, “Z” frame, and “+” frame. All these frames are computationally investigated with incorporate the following materials and their relevant mechanical properties: GFRP (Glass Fiber Reinforced Polymer), CFRP (Carbon Fiber Reinforced Polymer), KFRP (Kevlar Fiber Reinforced Polymer), and Aluminium Alloy. Apart from these fundamentals, the involvement of tools and their justifications are played a predominant role in this work. The major tools contributed are CATIA for modelling, ANSYS Workbench for Structural computations. The grid convergence test has been organized in a good manner to validate involved computational procedures. Finally, the suitable MR-UAV”s airframe and it opts material are obtained based on the multi-objective selection factors.

Pareto efficient buy and hold investment strategies under order book linked constraints

This paper deals with a multiobjective portfolio selection problem involving expected wealth (or return), coherent risk measures, deviation measures, and “level II order book data”, i.e., natural restrictions provoked by the existence of several levels of both bid and ask quotes with the corresponding depth. Obviously, the incorporation of the order book information makes our study much more realistic, since it is more closely related to the real behavior of many financial markets. Besides, ambiguity may be incorporated, which allows us to overcome the model-risk, since a potential discrepancy between the real (and maybe unknown) probabilities and the estimated ones is taken into account. Though the portfolio choice problem is not at all linear, its dual problem becomes a linear goal programming problem, and, consequently, the absence of duality gap allows us to solve the portfolio choice problem in an easy way. Furthermore, the double dual problem is linear too, and its solution also leads to the Pareto-efficient frontier. Lastly, we explore the influence on the Pareto-efficient frontier of the existence of “golden strategies”, i.e., investment strategies whose tail (or downside) risk is strictly lower than their price. Numerical experiments involving all the findings are implemented in a derivative market where both future contracts and call/put options may be traded.

Generating and Optimizing Human-Readable Quantitative Program Trading Strategies through a Genetic Programming Framework

In this paper, we provide a highly flexible genetic programming framework for automatic generation and optimization of program trading strategies. We propose the input/output modules and their implementation methods, decoupled from the GP kernel, making it a priori-posteriori framework for trading practitioners. For human-readable purposes, we also give various empirical regularization methods, including NSGA-II multi-objective selection, as well as experimentally effective performance measures.

A TOPSIS-Assisted Feature Selection Scheme and SOM-Based Anomaly Detection for Milling Tools Under Different Operating Conditions

Anomaly detection modeled as a one-class classification is an essential task for tool condition monitoring (TCM) when only the normal data are available. To confront with the real-world settings, it is crucial to take the different operating conditions, e.g., rotation speed, into account when approaching TCM solutions. This work mainly addresses issues related to multi-operating-condition TCM models, namely the varying discriminability of sensory features with different operating conditions; the overlap between normal and anomalous data; and the complex structure of input data. A feature selection scheme is proposed in which the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) is presented as a tool to aid the multi-objective selection of sensory features. In addition, four anomaly detection approaches based on Self-Organizing Map (SOM) are studied. To examine the stability of the four approaches, they are applied on different single-operating-condition models. Further, to examine their robustness when dealing with complex data structures, they are applied on multi-operating-condition models. The experimental results using the NASA Milling Data Set showed that all the studied anomaly detection approaches achieved a higher assessment accuracy with our feature selection scheme as compared to the Principal Component Analysis (PCA), Laplacian Score (LS), and extended LS in which we added a final step to the original LS method in order to eliminate redundant features.

Multi-objective selection of portfolios using ESG controversies

It is increasingly common for investors to demand a certain degree of compliance and commitment to environmental, social and governance (ESG) variables in their investments, without renouncing to maximising returns with the minimum possible risk. In this paper, a multi-objective optimisation model applied to Dow Jones stocks is used to analyse the construction of portfolios taking into account the level of controversies of each asset in the portfolio. Controversies are associated with non-compliance by companies in ESG areas, either due to a lack of dedication of resources or a lack of control. The a priori expected positive relationship between higher profitability and low number of controversies is not confirmed. This means that those companies with the highest returns are not necessarily those with the less controversies and vice versa. This finding contradicts the results obtained by some previous studies and underline the importance of applying multi-objective models to properly account for investors’ preferences, including ESG compliance.