Pub Date : 2024-10-22DOI: 10.1016/j.ins.2024.121582
The interpretability of disease prediction models is often crucial for their trustworthiness and usability among medical practitioners. Existing methods in interpretable artificial intelligence improve model transparency but fall short in identifying precise, disease-specific primal information. In this work, an interpretable deep learning-based algorithm called the data space landmark refiner was developed, which not only enhances both global interpretability and local interpretability but also reveals the intrinsic information of the data distribution. Using the proposed method, a type 2 diabetes mellitus diagnostic model with high interpretability was constructed on the basis of the electronic health records from two hospitals. Moreover, effective diagnostic information was directly derived from the model’s internal parameters, demonstrating strong alignment with current clinical knowledge. Compared with conventional interpretable machine learning approaches, the proposed method offered more precise and specific interpretability, increasing clinical practitioners’ trust in machine learning-supported diagnostic models.
{"title":"Unveiling diagnostic information for type 2 diabetes through interpretable machine learning","authors":"","doi":"10.1016/j.ins.2024.121582","DOIUrl":"10.1016/j.ins.2024.121582","url":null,"abstract":"<div><div>The interpretability of disease prediction models is often crucial for their trustworthiness and usability among medical practitioners. Existing methods in interpretable artificial intelligence improve model transparency but fall short in identifying precise, disease-specific primal information. In this work, an interpretable deep learning-based algorithm called the data space landmark refiner was developed, which not only enhances both global interpretability and local interpretability but also reveals the intrinsic information of the data distribution. Using the proposed method, a type 2 diabetes mellitus diagnostic model with high interpretability was constructed on the basis of the electronic health records from two hospitals. Moreover, effective diagnostic information was directly derived from the model’s internal parameters, demonstrating strong alignment with current clinical knowledge. Compared with conventional interpretable machine learning approaches, the proposed method offered more precise and specific interpretability, increasing clinical practitioners’ trust in machine learning-supported diagnostic models.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":null,"pages":null},"PeriodicalIF":8.1,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142530661","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-22DOI: 10.1016/j.ins.2024.121577
Belief function (BF) theory provides a framework for effective modeling, quantifying uncertainty, and combining evidence, rendering it a potent tool for tackling uncertain decision-making problems. However, with the expansion of the frame of discernment, the increasing number of focal elements processed during the fusion procedure leads to a rapid increase in computational complexity, which limits the practical application of BF theory. To overcome this issue, a novel multi-granularity belief structure (MGBS) method was proposed in this study. The construction of MGBS reduced the number of focal elements and preserved crucial information in the basic belief assignment. This effectively reduced the computational complexity of fusion while ensuring the highest possible classification accuracy. We applied the proposed MGBS algorithm to a human activity recognition task and verified its effectiveness using the University of California, Irvine mHealth, PAMAP2, and Smartphone datasets.
{"title":"Evidence combination with multi-granularity belief structure for pattern classification","authors":"","doi":"10.1016/j.ins.2024.121577","DOIUrl":"10.1016/j.ins.2024.121577","url":null,"abstract":"<div><div>Belief function (BF) theory provides a framework for effective modeling, quantifying uncertainty, and combining evidence, rendering it a potent tool for tackling uncertain decision-making problems. However, with the expansion of the frame of discernment, the increasing number of focal elements processed during the fusion procedure leads to a rapid increase in computational complexity, which limits the practical application of BF theory. To overcome this issue, a novel multi-granularity belief structure (MGBS) method was proposed in this study. The construction of MGBS reduced the number of focal elements and preserved crucial information in the basic belief assignment. This effectively reduced the computational complexity of fusion while ensuring the highest possible classification accuracy. We applied the proposed MGBS algorithm to a human activity recognition task and verified its effectiveness using the University of California, Irvine mHealth, PAMAP2, and Smartphone datasets.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":null,"pages":null},"PeriodicalIF":8.1,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142530663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-22DOI: 10.1016/j.ins.2024.121578
Axiomatic fully probabilistic design (FPD) of optimal decision rules strictly extends the decision making (DM) theory represented by Markov decision processes (MDP). This means that any MDP task can be approximated by an explicitly found FPD task whereas many FPD tasks have no MDP equivalent. MDP and FPD model the closed loop — the coupling of an agent and its environment — via a joint probability density (pd) relating the involved random variables, referred to as behaviour. Unlike MDP, FPD quantifies agent's aims and constraints by an ideal pd. The ideal pd is high on the desired behaviours, small on undesired behaviours and zero on forbidden ones. FPD selects the optimal decision rules as the minimiser of Kullback-Leibler's divergence of the closed-loop-modelling pd to its ideal twin. The proximity measure choice follows from the FPD axiomatics.
MDP minimises the expected total loss, which is usually the sum of discounted partial losses. The discounting reflects the decreasing importance of future losses. It also diminishes the influence of errors caused by:
the imperfection of the employed environment model;
roughly-expressed aims;
the approximate learning and decision-rules design.
The established FPD cannot currently account for these important features. The paper elaborates the missing discounted version of FPD. This non-trivial filling of the gap in FPD also employs an extension of dynamic programming, which is of an independent interest.
{"title":"Discounted fully probabilistic design of decision rules","authors":"","doi":"10.1016/j.ins.2024.121578","DOIUrl":"10.1016/j.ins.2024.121578","url":null,"abstract":"<div><div>Axiomatic fully probabilistic design (FPD) of optimal decision rules strictly extends the decision making (DM) theory represented by Markov decision processes (MDP). This means that any MDP task can be approximated by an explicitly found FPD task whereas many FPD tasks have no MDP equivalent. MDP and FPD model the closed loop — the coupling of an agent and its environment — via a joint probability density (pd) relating the involved random variables, referred to as behaviour. Unlike MDP, FPD quantifies agent's aims and constraints by an <em>ideal pd</em>. The ideal pd is high on the desired behaviours, small on undesired behaviours and zero on forbidden ones. FPD selects the optimal decision rules as the minimiser of Kullback-Leibler's divergence of the closed-loop-modelling pd to its ideal twin. The proximity measure choice follows from the FPD axiomatics.</div><div>MDP minimises the expected total loss, which is usually the sum of discounted partial losses. The discounting reflects the decreasing importance of future losses. It also diminishes the influence of errors caused by:</div><div><figure><img></figure> the imperfection of the employed environment model;</div><div><figure><img></figure> roughly-expressed aims;</div><div><figure><img></figure> the approximate learning and decision-rules design.</div><div>The established FPD cannot currently account for these important features. The paper elaborates the missing discounted version of FPD. This non-trivial filling of the gap in FPD also employs an extension of dynamic programming, which is of an independent interest.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":null,"pages":null},"PeriodicalIF":8.1,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142530657","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-22DOI: 10.1016/j.ins.2024.121573
The decomposition of all pseudo-uninorms with continuous underlying functions, defined on the unit interval, via Clifford's ordinal sum is described. It is shown that each such pseudo-uninorm can be decomposed into representable and trivial semigroups, and special semigroups defined on two points, where the corresponding semigroup operation is the projection to one of the coordinates. Linear orders, for which the ordinal sum of such semigroups yields a pseudo-uninorm, are also characterized.
{"title":"Decomposition of pseudo-uninorms with continuous underlying functions via ordinal sum","authors":"","doi":"10.1016/j.ins.2024.121573","DOIUrl":"10.1016/j.ins.2024.121573","url":null,"abstract":"<div><div>The decomposition of all pseudo-uninorms with continuous underlying functions, defined on the unit interval, via Clifford's ordinal sum is described. It is shown that each such pseudo-uninorm can be decomposed into representable and trivial semigroups, and special semigroups defined on two points, where the corresponding semigroup operation is the projection to one of the coordinates. Linear orders, for which the ordinal sum of such semigroups yields a pseudo-uninorm, are also characterized.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":null,"pages":null},"PeriodicalIF":8.1,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142530660","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-22DOI: 10.1016/j.ins.2024.121585
Dimensionality reduction is a critical factor in processing high-dimensional datasets. The L1 norm-based Two-Dimensional Linear Discriminant Analysis (L1-2DLDA) is widely used for this purpose, but it remains sensitive to outliers and classes with large deviations, which deteriorates its performance. To address this limitation, the present study proposed Pairwise Sample Distance Two-Dimensional Linear Discriminant Analysis (PSD2DLDA), a novel method that modeled L1-2DLDA using pair-wise sample distances. To improve computational effectiveness, this study also introduced a streamlined variant, Pairwise Class Mean Distance Two-Dimensional Linear Discriminant Analysis (PCD2DLDA), which was based on distances between class mean pairs. Different from previous studies, this study utilized the projected sub-gradient method to optimize these two improved methods. Meanwhile, this study explored the interrelationship, limitations, and applicability of these two improved methods. The comparative experimental results on three datasets validated the outstanding performance of PSD2DLDA and PCD2DLDA methods. In particular, PSD2DLDA exhibited superior robustness compared to PCD2DLDA. Furthermore, applying these two methods to optimize electroencephalogram (EEG) signals effectively enhanced the decoding accuracy of motor imagery neural patterns, which offered a promising strategy for optimizing EEG signals processing in brain-computer interface (BCI) applications.
{"title":"Improving two-dimensional linear discriminant analysis with L1 norm for optimizing EEG signal","authors":"","doi":"10.1016/j.ins.2024.121585","DOIUrl":"10.1016/j.ins.2024.121585","url":null,"abstract":"<div><div>Dimensionality reduction is a critical factor in processing high-dimensional datasets. The L1 norm-based Two-Dimensional Linear Discriminant Analysis (L1-2DLDA) is widely used for this purpose, but it remains sensitive to outliers and classes with large deviations, which deteriorates its performance. To address this limitation, the present study proposed Pairwise Sample Distance Two-Dimensional Linear Discriminant Analysis (PSD2DLDA), a novel method that modeled L1-2DLDA using pair-wise sample distances. To improve computational effectiveness, this study also introduced a streamlined variant, Pairwise Class Mean Distance Two-Dimensional Linear Discriminant Analysis (PCD2DLDA), which was based on distances between class mean pairs. Different from previous studies, this study utilized the projected sub-gradient method to optimize these two improved methods. Meanwhile, this study explored the interrelationship, limitations, and applicability of these two improved methods. The comparative experimental results on three datasets validated the outstanding performance of PSD2DLDA and PCD2DLDA methods. In particular, PSD2DLDA exhibited superior robustness compared to PCD2DLDA. Furthermore, applying these two methods to optimize electroencephalogram (EEG) signals effectively enhanced the decoding accuracy of motor imagery neural patterns, which offered a promising strategy for optimizing EEG signals processing in brain-computer interface (BCI) applications.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":null,"pages":null},"PeriodicalIF":8.1,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142538694","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-22DOI: 10.1016/j.ins.2024.121551
To enhance the conventional framework of data envelope analysis (DEA), a novel hybrid bi-level model is proposed, integrating fuzzy logic with triangular fuzzy numbers to effectively address data uncertainty. This model innovatively departs from the traditional DEA’s ’black box’ approach by incorporating inter-organizational relationships and the internal dynamics of decision-making units (DMUs). Utilizing a modified Russell’s method, it provides a nuanced efficiency analysis in scenarios of ambiguous data. The study aims to enhance the accuracy and applicability of Data Envelopment Analysis in uncertain data environments. To achieve this, a novel hybrid bi-level model integrating fuzzy logic is presented. Validated through a case study involving 15 branches of a private Iranian bank, the model demonstrates improved accuracy in efficiency assessments and paves the way for future research in operational systems uncertainty management. The results indicated that, among the 15 branches of a private Iranian bank analyzed for the year 2022, branches 1, 10, and 11 demonstrated leader-level efficiency, while branch 3 exhibited follower-level efficiency, and branch 1 achieved overall efficiency. These branches attained an efficiency rating of , signifying a high level of efficiency within the model’s parameters.
{"title":"Efficiency analysis in bi-level on fuzzy input and output","authors":"","doi":"10.1016/j.ins.2024.121551","DOIUrl":"10.1016/j.ins.2024.121551","url":null,"abstract":"<div><div>To enhance the conventional framework of data envelope analysis (DEA), a novel hybrid bi-level model is proposed, integrating fuzzy logic with triangular fuzzy numbers to effectively address data uncertainty. This model innovatively departs from the traditional DEA’s ’black box’ approach by incorporating inter-organizational relationships and the internal dynamics of decision-making units (DMUs). Utilizing a modified Russell’s method, it provides a nuanced efficiency analysis in scenarios of ambiguous data. The study aims to enhance the accuracy and applicability of Data Envelopment Analysis in uncertain data environments. To achieve this, a novel hybrid bi-level model integrating fuzzy logic is presented. Validated through a case study involving 15 branches of a private Iranian bank, the model demonstrates improved accuracy in efficiency assessments and paves the way for future research in operational systems uncertainty management. The results indicated that, among the 15 branches of a private Iranian bank analyzed for the year 2022, branches 1, 10, and 11 demonstrated leader-level efficiency, while branch 3 exhibited follower-level efficiency, and branch 1 achieved overall efficiency. These branches attained an efficiency rating of <span><math><mrow><msup><mi>E</mi><mrow><mo>+</mo><mo>+</mo></mrow></msup></mrow></math></span>, signifying a high level of efficiency within the model’s parameters.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":null,"pages":null},"PeriodicalIF":8.1,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142530615","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-22DOI: 10.1016/j.ins.2024.121574
Variable selection is important for classification of data with many irrelevant predicting variables, but it has not yet been well studied in positive-unlabeled (PU) learning, where classifiers have to be trained without labelled-negative instances. In this paper, we propose a group kernel-free PU classifier with asymmetric loss (GKF-PUAL) to achieve quadratic PU classification with group-lasso regularisation embedded for variable selection. We also propose a five-block algorithm to solve the optimization problem of GKF-PUAL. Our experimental results reveal the superiority of GKF-PUAL in both PU classification and variable selection, improving the baseline PUAL by more than 10% in F1-score across four benchmark datasets and removing over 70% of irrelevant variables on six benchmark datasets. The code for GKF-PUAL is at https://github.com/tkks22123/GKF-PUAL.
变量选择对于具有许多不相关预测变量的数据分类非常重要，但在正向无标记（PU）学习中还没有得到很好的研究，在这种学习中，分类器必须在没有标记负实例的情况下进行训练。在本文中，我们提出了一种具有非对称损失的无组核 PU 分类器（GKF-PUAL），通过嵌入用于变量选择的组-拉索正则化来实现二次 PU 分类。我们还提出了一种五块算法来解决 GKF-PUAL 的优化问题。我们的实验结果表明，GKF-PUAL 在 PU 分类和变量选择方面都具有优越性，在四个基准数据集上的 F1 分数比基准 PUAL 提高了 10%以上，并在六个基准数据集上去除了 70% 以上的无关变量。GKF-PUAL 的代码见 https://github.com/tkks22123/GKF-PUAL。
{"title":"GKF-PUAL: A group kernel-free approach to positive-unlabeled learning with variable selection","authors":"","doi":"10.1016/j.ins.2024.121574","DOIUrl":"10.1016/j.ins.2024.121574","url":null,"abstract":"<div><div>Variable selection is important for classification of data with many irrelevant predicting variables, but it has not yet been well studied in positive-unlabeled (PU) learning, where classifiers have to be trained without labelled-negative instances. In this paper, we propose a group kernel-free PU classifier with asymmetric loss (GKF-PUAL) to achieve quadratic PU classification with group-lasso regularisation embedded for variable selection. We also propose a five-block algorithm to solve the optimization problem of GKF-PUAL. Our experimental results reveal the superiority of GKF-PUAL in both PU classification and variable selection, improving the baseline PUAL by more than 10% in F1-score across four benchmark datasets and removing over 70% of irrelevant variables on six benchmark datasets. The code for GKF-PUAL is at <span><span>https://github.com/tkks22123/GKF-PUAL</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":null,"pages":null},"PeriodicalIF":8.1,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142530656","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-22DOI: 10.1016/j.ins.2024.121571
Multi-Constrained Graph Pattern Matching (MC-GPM) aims to match a pattern graph with multiple attribute constraints on its nodes and edges, and has garnered significant interest in various fields, including social-based e-commerce and trust-based group discovery. However, the existing MC-GPM methods do not consider situations where the number of each node in the pattern graph needs to be fixed, such as finding experts group with expert quantities and relations specified. In this paper, a Multi-Constrained Strong Simulation with the Fixed Number of Nodes (MCSS-FNN) matching model is proposed, and then a Trust-oriented Optimal Multi-constrained Path (TOMP) matching algorithm is designed for solving it. Additionally, two heuristic optimization strategies are designed, one for combinatorial testing and the other for edge matching, to enhance the efficiency of the TOMP algorithm. Empirical experiments are conducted on four real social network datasets, and the results demonstrate the effectiveness and efficiency of the proposed algorithm and optimization strategies.
{"title":"Size-fixed group discovery via multi-constrained graph pattern matching","authors":"","doi":"10.1016/j.ins.2024.121571","DOIUrl":"10.1016/j.ins.2024.121571","url":null,"abstract":"<div><div>Multi-Constrained Graph Pattern Matching (MC-GPM) aims to match a pattern graph with multiple attribute constraints on its nodes and edges, and has garnered significant interest in various fields, including social-based e-commerce and trust-based group discovery. However, the existing MC-GPM methods do not consider situations where the number of each node in the pattern graph needs to be fixed, such as finding experts group with expert quantities and relations specified. In this paper, a Multi-Constrained Strong Simulation with the Fixed Number of Nodes (MCSS-FNN) matching model is proposed, and then a Trust-oriented Optimal Multi-constrained Path (TOMP) matching algorithm is designed for solving it. Additionally, two heuristic optimization strategies are designed, one for combinatorial testing and the other for edge matching, to enhance the efficiency of the TOMP algorithm. Empirical experiments are conducted on four real social network datasets, and the results demonstrate the effectiveness and efficiency of the proposed algorithm and optimization strategies.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":null,"pages":null},"PeriodicalIF":8.1,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142530664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-22DOI: 10.1016/j.ins.2024.121569
In the real world, the decision variables of large-scale sparse multi-objective problems are high-dimensional, and most Pareto optimal solutions are sparse. The balance of the algorithms is difficult to control, so it is challenging to deal with such problems in general. Therefore, An Enhanced Competitive Swarm Optimizer with Strongly Robust Sparse Operator (SR-ECSO) algorithm is proposed. Firstly, the strongly robust sparse functions which accelerate particles in the population better sparsity in decision space, are used in high-dimensional decision variables. Secondly, the diversity of sparse solutions is maintained, and the convergence balance of the algorithm is enhanced by the introduction of an adaptive random perturbation operator. Finally, the state of the particles is updated using a swarm optimizer to improve population competitiveness. To verify the proposed algorithm, we tested eight large-scale sparse benchmark problems, and the decision variables were set in three groups with 100, 500, and 1000 as examples. Experimental results show that the algorithm is promising for solving large-scale sparse optimization problems.
{"title":"An enhanced competitive swarm optimizer with strongly robust sparse operator for large-scale sparse multi-objective optimization problem","authors":"","doi":"10.1016/j.ins.2024.121569","DOIUrl":"10.1016/j.ins.2024.121569","url":null,"abstract":"<div><div>In the real world, the decision variables of large-scale sparse multi-objective problems are high-dimensional, and most Pareto optimal solutions are sparse. The balance of the algorithms is difficult to control, so it is challenging to deal with such problems in general. Therefore, An Enhanced Competitive Swarm Optimizer with Strongly Robust Sparse Operator (SR-ECSO) algorithm is proposed. Firstly, the strongly robust sparse functions which accelerate particles in the population better sparsity in decision space, are used in high-dimensional decision variables. Secondly, the diversity of sparse solutions is maintained, and the convergence balance of the algorithm is enhanced by the introduction of an adaptive random perturbation operator. Finally, the state of the particles is updated using a swarm optimizer to improve population competitiveness. To verify the proposed algorithm, we tested eight large-scale sparse benchmark problems, and the decision variables were set in three groups with 100, 500, and 1000 as examples. Experimental results show that the algorithm is promising for solving large-scale sparse optimization problems.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":null,"pages":null},"PeriodicalIF":8.1,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142530665","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-22DOI: 10.1016/j.ins.2024.121572
Multi-objective evolutionary algorithms have shown their competitiveness in solving ROC convex hull maximization. However, due to “the curse of dimensionality”, few of them focus on high-dimensional ROCCH maximization. Therefore, in this paper, a feedback matrix (FM)-based evolutionary multitasking algorithm, termed as FM-EMTA, is proposed. In FM-EMTA, to tackle “the curse of dimensionality”, a feature importance based low-dimensional task construction strategy is designed to transform the high-dimensional ROCCH maximization task into several low-dimensional tasks. Then, each low-dimensional task evolves with a population. To ensure that the low-dimensional task achieves a better ROCCH, an FM-based evolutionary multitasking operator is proposed. Specifically, for each low-dimensional task i, the element FM(i,j) in feedback matrix is defined to measure the degree that the low-dimensional task j could assist task i. Based on it, an FM-based assisted task selection operator and an FM-based knowledge transfer operator are developed to constitute the evolutionary multitasking operator, with which the useful knowledge is transferred among the low-dimensional tasks. After the evolution, the best ROCCHs obtained by the low-dimensional tasks are combined together to achieve the final ROCCH on the original high-dimensional task. Experiments on twelve high-dimensional datasets with different characteristics demonstrate the superiority of the proposed FM-EMTA over the state-of-the-arts in terms of the area under ROCCH, the hypervolume indicator and the running time.
{"title":"A feedback matrix based evolutionary multitasking algorithm for high-dimensional ROC convex hull maximization","authors":"","doi":"10.1016/j.ins.2024.121572","DOIUrl":"10.1016/j.ins.2024.121572","url":null,"abstract":"<div><div>Multi-objective evolutionary algorithms have shown their competitiveness in solving ROC convex hull maximization. However, due to “the curse of dimensionality”, few of them focus on high-dimensional ROCCH maximization. Therefore, in this paper, a feedback matrix (<strong>FM</strong>)-based evolutionary multitasking algorithm, termed as FM-EMTA, is proposed. In FM-EMTA, to tackle “the curse of dimensionality”, a feature importance based low-dimensional task construction strategy is designed to transform the high-dimensional ROCCH maximization task into several low-dimensional tasks. Then, each low-dimensional task evolves with a population. To ensure that the low-dimensional task achieves a better ROCCH, an <strong>FM</strong>-based evolutionary multitasking operator is proposed. Specifically, for each low-dimensional task <em>i</em>, the element <strong>FM</strong>(<em>i</em>,<em>j</em>) in feedback matrix is defined to measure the degree that the low-dimensional task <em>j</em> could assist task <em>i</em>. Based on it, an <strong>FM</strong>-based assisted task selection operator and an <strong>FM</strong>-based knowledge transfer operator are developed to constitute the evolutionary multitasking operator, with which the useful knowledge is transferred among the low-dimensional tasks. After the evolution, the best ROCCHs obtained by the low-dimensional tasks are combined together to achieve the final ROCCH on the original high-dimensional task. Experiments on twelve high-dimensional datasets with different characteristics demonstrate the superiority of the proposed FM-EMTA over the state-of-the-arts in terms of the area under ROCCH, the hypervolume indicator and the running time.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":null,"pages":null},"PeriodicalIF":8.1,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142530658","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号