Article
Management
Hassan T. Anis, Roy H. Kwon
Summary: This paper investigates the cardinality constrained risk parity optimization problem and proposes two formulations that can be solved to global optimality using existing solvers. The experiments show that the convex formulation is efficient in terms of both speed and accuracy, and the resulting portfolios exhibit great out-of-sample performance.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2022)
Article
Computer Science, Artificial Intelligence
Man-Fai Leung, Jun Wang
Summary: This paper presents a collaborative neurodynamic optimization approach for cardinality-constrained portfolio selection, solving the mixed-integer optimization problem using multiple recurrent neural networks and particle swarm optimization. Experimental results demonstrate the superior performance of this approach in handling stock data compared to other methods.
Article
Mathematics
Francisco Fernandez-Navarro, Luisa Martinez-Nieto, Mariano Carbonero-Ruz, Teresa Montero-Romero
Summary: This paper introduces the mean-variance (MV) portfolio and mean squared variance (MSV) portfolio methods, and proposes a mixed-integer linear programming (MILP) reformation for the non-convex QP problem, as well as a data-driven method for determining the optimal value of the hyper-parameter. Empirical tests show that the MSV portfolio exhibits competitive performance in most problems.
Article
Energy & Fuels
Alexander Thebelt, Calvin Tsay, Robert M. Lee, Nathan Sudermann-Merx, David Walz, Tom Tranter, Ruth Misener
Summary: This paper proposes a novel data-driven strategy using tree ensembles for constrained multi-objective optimization of black-box problems with heterogeneous variable spaces. It demonstrates competitive performance and sampling efficiency in real-world applications with limited evaluation budgets compared to other state-of-the-art tools.
Article
Operations Research & Management Science
Moritz Link, Stefan Volkwein
Summary: In this paper, a new method is proposed for computing an enclosure of the nondominated set of multiobjective mixed-integer quadratically constrained programs without any convexity requirements. The method uses piecewise linear relaxations to bypass the nonconvexity of the original problem. It adaptsively chooses the level of relaxation needed in different parts of the image space. After finitely many iterations, an enclosure of the nondominated set of prescribed quality is guaranteed. The advantages of this approach are demonstrated through its application to multiobjective energy supply network problems.
JOURNAL OF GLOBAL OPTIMIZATION
(2023)
Article
Computer Science, Artificial Intelligence
Meriem Hemici, Djaafar Zouache
Summary: This paper proposes a new multi-objective evolutionary algorithm called MP-MOEA, which is based on multi-population, to solve the multi-objective constrained portfolio optimization problem in finance. By using a multi-population strategy and two types of archives, the algorithm improves solution quality, accelerates convergence, and demonstrates superior performance in experiments.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Management
Ilgin Dogan, Banu Lokman, Murat Koksalan
Summary: This paper discusses generating a representative subset of nondominated points in multi-objective mixed-integer programs and develops an exact algorithm to guarantee a prespecified precision. Experiments demonstrate that the algorithm outperforms existing approaches in terms of both the cardinality of the representative set and computation times.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2022)
Article
Operations Research & Management Science
Tommaso Giovannelli, Giampaolo Liuzzi, Stefano Lucidi, Francesco Rinaldi
Summary: This paper focuses on mixed-integer nonsmooth constrained optimization problems where objective/constraint functions are only available as the output of a black-box zeroth-order oracle that lacks derivative information. The authors propose a novel derivative-free linesearch-based algorithmic framework to handle these problems effectively. They first describe a scheme for bound constrained problems that combines dense sequence directions with primitive directions to handle the nonsmoothness of the objective function and discrete variables. Then, they embed an exact penalty approach in the scheme to manage nonlinear (possibly nonsmooth) constraints suitably. The proposed algorithms are analyzed for their global convergence properties towards stationary points, and extensive numerical experiments on a set of mixed-integer test problems are presented.
COMPUTATIONAL OPTIMIZATION AND APPLICATIONS
(2022)
Article
Thermodynamics
Miguel Gonzalez-Salazar, Julia Klossek, Pascal Dubucq, Thomas Punde
Summary: Long-term portfolio optimization for district heating systems is challenging due to the need for high accuracy and computational speed. This paper investigates the advantages and disadvantages of using merit order (MO) models compared to mixed integer linear programming (MILP) models. Results suggest that MO models, especially those incorporating heat storage and detailed description of CHP plants, can significantly reduce computation time without sacrificing accuracy. Combining MO and MILP models offers a faster and more robust decision-making process.
Article
Management
Fenlan Wang
Summary: This paper proposes a new method for solving integer quadratic programming problems by exploiting the characteristics of quadratic contours and the integer points of the continuous optimal solution, cutting off sub-boxes that do not contain the optimal solution. By integrating this solution scheme into a branch-and-bound algorithm, the proposed method reduces the optimality gap successively in the solution iterations and has a finite-step convergence.
JOURNAL OF THE OPERATIONAL RESEARCH SOCIETY
(2021)
Article
Computer Science, Artificial Intelligence
Hong Zhao, Zong-Gan Chen, Zhi-Hui Zhan, Sam Kwong, Jun Zhang
Summary: This paper investigates the multi-objective cardinality constrained portfolio optimization problem and proposes a multiple populations co-evolutionary particle swarm optimization algorithm to address this issue. The algorithm has advantages in dealing with cardinality constraints and multi-objective challenges through strategies such as hybrid encoding, heuristic method, local search, and elite competition.
Article
Mathematics
Loay Alkhalifa, Hans Mittelmann
Summary: This paper introduces the method of piecewise linear approximation (PLA) and its application on mixed integer nonlinear programming (MINLP) problems. By using nonuniform domain partitioning, the PLA models are improved to obtain more accurate solutions and reduce computation time.
Article
Computer Science, Artificial Intelligence
Yongkuan Yang, Jianchang Liu, Shubin Tan
Summary: The paper introduces a multi objective differential evolutionary algorithm based on partition selection (MODE-PS) to tackle constrained multi-objective optimization problems. By dividing problems into sub-spaces and maintaining feasibility, the algorithm accelerates convergence and proves to be competitive in solving CMOPs.
SWARM AND EVOLUTIONARY COMPUTATION
(2021)
Article
Computer Science, Information Systems
Xue Feng, Anqi Pan, Zhengyun Ren, Zhiping Fan
Summary: Balancing convergence and diversity is a challenge in multi-objective optimization problems, especially when the proportion of feasible regions is low. This paper proposes a constrained multi-objective optimization algorithm based on a hybrid driven strategy to enhance the feasibility and diversity performance of Pareto solutions. The algorithm outperforms peer algorithms, especially in large-infeasible-regions multi-objective optimization problems.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Yongkuan Yang, Pei-Qiu Huang, Xiangsong Kong, Jing Zhao
Summary: This paper proposes a novel constrained multi-objective evolutionary algorithm called CMAOO, which optimizes an (M+1)-objective optimization problem consisting of the original M objective functions and the degree of constraint violation. It constructs a main population and saves all feasible solutions in an external archive. The main population and the external archive are evolved to search the whole space and the feasible regions, respectively, and their offspring update the external archive and the main population separately. Experimental studies show that CMAOO is competitive in solving constrained multi-objective optimization problems compared to four state-of-the-art algorithms.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Quoc Bao Diep, Thanh Cong Truong, Swagatam Das, Ivan Zelinka
Summary: This article introduces an improved version of the Self-Organizing Migrating Algorithm named iSOMA and evaluates its performance. The iSOMA algorithm shows notable improvements compared to previous versions and achieves excellent results in multiple benchmark tests. Additionally, the article demonstrates the application of iSOMA in drone path planning.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Arka Ghosh, Sankha Subhra Mullick, Shounak Datta, Swagatam Das, Asit Kr Das, Rammohan Mallipeddi
Summary: This study introduces the DEceit algorithm, which constructs effective universal pixel-restricted perturbations using only black-box feedback from the target network. Through empirical investigations, it is found that perturbing around 10% of the pixels in an image achieves a highly transferable Fooling Rate while maintaining visual quality. Furthermore, DEceit shows success in image-dependent attacks and outperforms several state-of-the-art methods.
PATTERN RECOGNITION
(2022)
Article
Computer Science, Artificial Intelligence
Souhardya Sengupta, Swagatam Das
Summary: This paper proposes a simple data clustering technique that uses a graph created on nearest neighbors to identify clusters. The algorithm incorporates border detection and outlier detection techniques to construct the graph. The authors also introduce a novel outlier detection technique suitable for their implementation. Comparisons with state-of-the-art clustering techniques and experiments on various aspects of datasets are conducted.
PATTERN RECOGNITION LETTERS
(2022)
Article
Energy & Fuels
Oladayo S. Ajani, Abhishek Kumar, Rammohan Mallipeddi, Swagatam Das, Ponnuthurai Nagaratnam Suganthan
Summary: Energy disaggregation is an effective method to promote energy efficiency, but it faces challenges such as device similarity and measurement errors. In order to develop optimization algorithms for energy disaggregation, standard datasets and evaluation metrics are needed. This paper proposes a dataset with multiple instances and summarizes the performance indicators and baseline results of different optimization algorithms.
Article
Automation & Control Systems
Pourya Shamsolmoali, Masoumeh Zareapoor, Swagatam Das, Salvador Garcia, Eric Granger, Jie Yang
Summary: Image-to-image translation is crucial in generative adversarial networks. Convolutional neural networks have limitations in capturing spatial relationships, making them unsuitable for image translation tasks. Capsule networks are proposed as a remedy, capturing hierarchical spatial relationships. In this paper, a new framework for capsule networks is presented, which can be applied to generator-discriminator architectures without computational overhead. A Gromov-Wasserstein distance is used as a loss function to guide the learned distribution. The proposed method, called generative equivariant network, is evaluated on I2I translation and image generation tasks and shows a principled connection between generative and capsule models.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Review
Computer Science, Artificial Intelligence
Kanchan Rajwar, Kusum Deep, Swagatam Das
Summary: As industrialization progresses, solving optimization problems becomes more challenging. More than 500 new metaheuristic algorithms (MAs) have been developed, with over 350 of them emerging in the last decade. This study tracks approximately 540 MAs and provides statistical information. The proliferation of MAs has led to the issue of significant similarities between algorithms with different names. The study categorizes MAs based on the number of control parameters, which is a new taxonomy. Real-world applications of MAs are demonstrated and limitations and open challenges are identified.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Automation & Control Systems
Saptarshi Chakraborty, Debolina Paul, Swagatam Das
Summary: This article introduces a center-based clustering method that incorporates an entropy incentive term to learn feature importance efficiently. A scalable block-coordinate descent algorithm with closed-form updates is used to minimize the objective function. The merits of this method are showcased through detailed experimental analysis.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Computer Science, Artificial Intelligence
Kushal Bose, Swagatam Das
Summary: Graph Neural Networks (GNNs) are powerful for learning on graph-structured data, but they often suffer from over-smoothing, where node features become indistinguishable. This paper identifies the recursive and higher-to-lower order aggregation as the primary causes of over-smoothing and proposes a novel non-recursive aggregation strategy using randomized path exploration. Extensive comparative studies on benchmark datasets demonstrate the efficacy of the proposed method in semi-supervised and fully-supervised learning tasks.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Debolina Paul, Saptarshi Chakraborty, Swagatam Das
Summary: Principal component analysis (PCA) is widely used for data visualization, denoising, and dimensionality reduction, but it is sensitive to outliers and may fail to detect low-dimensional structures. This article proposes a PCA method called MoMPCA based on the Median of Means (MoM) principle, which is computationally efficient and achieves optimal convergence rates. The method is robust to outliers and does not make assumptions about them. The efficacy of the proposed method is demonstrated through simulations and real data applications.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Review
Computer Science, Information Systems
Arkaprabha Basu, Sandip Paul, Sreeya Ghosh, Swagatam Das, Bhabatosh Chanda, Chakravarthy Bhagvati, Vaclav Snasel
Summary: Digitized methodologies have revolutionized various fields, including the restoration of buildings with historical significance. This interdisciplinary field attracts computer scientists who use computerized tools to reconstruct the values of these structures. The wear of time has endangered significant historical values, but this survey explores the use of 3D reconstruction, image inpainting, IoT-based methods, genetic algorithms, and image processing to restore cultural heritage. Machine Learning, Deep Learning, and Computer Vision-based methods are discussed, offering insights into faster, cheaper, and more beneficial techniques for image reconstruction in the near future.
Article
Computer Science, Artificial Intelligence
Debolina Paul, Saptarshi Chakraborty, Swagatam Das, Jason Xu
Summary: Kernel k-means clustering is a powerful tool for unsupervised learning of non-linearly separable data. In this paper, we propose a novel algorithm called Kernel Power k-Means, which leverages a general family of means to combat sub-optimal local solutions in the kernel and multi-kernel settings. Our algorithm uses majorization-minimization to solve the non-convex problem and implicitly performs annealing in kernel feature space. We rigorously analyze its convergence properties and demonstrate its efficacy through various experiments.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Information Systems
Arjun Ghosh, Nanda Dulal Jana, Swagatam Das, Rammohan Mallipeddi
Summary: This study proposes a two-phase evolutionary framework, TPEvo-CNN, for automatically designing CNN models for medical image classification. The framework utilizes differential evolution to determine the number of layers of the CNN architecture and genetic algorithm to fine-tune the hyperparameters. Experimental results demonstrate the superiority of the proposed framework in medical image classification tasks compared to existing CNN models.
Proceedings Paper
Computer Science, Artificial Intelligence
Lingping Kong, Vaclav Snasel, Swagatam Das, Jeng-Shyang Pan
Summary: This paper introduces a sorting method based on Location Gradient (LG) number, which is particularly efficient in dealing with duplicate solutions and can achieve a time complexity of O(N log N) in many cases, but still costs O(MN2) in the worst case scenario. The efficacy of the LG-based sorting method is demonstrated by comparing it against several existing non-dominated sorting procedures.
ADVANCES IN INTELLIGENT SYSTEMS AND COMPUTING (ECC 2021)
(2022)
Article
Computer Science, Artificial Intelligence
Avisek Gupta, Swagatam Das
Summary: Recent studies have shown that multiple kernel clustering methods have been highly successful in multi-view clustering of complex datasets. The proposed MKTC method simultaneously learns a multiple kernel metric and transfers it in two tasks for clustering.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Ivan Zelinka, Swagatam Das
Summary: This paper presents an approach to solve a variant of TSP using a gamesourcing method. By simulating ant colony optimization in a maze game, the TSP problem can be effectively solved.
Article
Computer Science, Artificial Intelligence
Guiliang Gong, Jiuqiang Tang, Dan Huang, Qiang Luo, Kaikai Zhu, Ningtao Peng
Summary: This paper proposes a flexible job shop scheduling problem with discrete operation sequence flexibility and designs an improved memetic algorithm to solve it. Experimental results show that the algorithm outperforms other algorithms in terms of performance. The proposed model and algorithm can help production managers obtain optimal scheduling schemes considering operations with or without sequence constraints.
SWARM AND EVOLUTIONARY COMPUTATION
(2024)
Article
Computer Science, Artificial Intelligence
Daniel Molina-Perez, Efren Mezura-Montes, Edgar Alfredo Portilla-Flores, Eduardo Vega-Alvarado, Barbara Calva-Yanez
Summary: This paper presents a new proposal based on two fundamental strategies to improve the performance of the differential evolution algorithm when solving MINLP problems. The proposal considers a set of good fitness-infeasible solutions to explore promising regions and introduces a composite trial vector generation method to enhance combinatorial exploration and convergence capacity.
SWARM AND EVOLUTIONARY COMPUTATION
(2024)
