Article
Multidisciplinary Sciences
Xiaoyu Tang, Sijia Xu, Hui Ye
Summary: This paper studies the problem of optimal matching in financial investment using machine learning algorithms, aiming to reduce the impact of information asymmetry for investors. By comparing various models using financial data, the optimal prediction model is determined, and concepts such as mean-variance model, Sharpe ratio, and efficient frontier are introduced to find a balance between risk and return. The study also analyzes the impact of investor behavior on trading strategies.
Article
Computer Science, Artificial Intelligence
Yinnan Chen, Xinchao Zhao, Junling Hao
Summary: In this paper, a new three-objective Sortino ratio-expected shortfall-turnover rate (SR-ES-TR) portfolio optimization (PO) model is proposed, which effectively manages investment risk and obtains considerable returns.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Green & Sustainable Science & Technology
Sasan Barak, Reza Moghdani, Hamidreza Maghsoudlou
Summary: This paper proposes a novel scheduling approach for a resource-constrained Flexible Manufacturing System (FMS) by considering energy efficiency of AGVs and using a modified multi-objective particle swarm optimization algorithm to solve the problem, which outperforms the classic version of the algorithm according to the results.
JOURNAL OF CLEANER PRODUCTION
(2021)
Article
Automation & Control Systems
Li Li, Liang Chang, Tianlong Gu, Weiguo Sheng, Wanliang Wang
Summary: This paper introduces a novel multiobjective PSO algorithm named MOPSO/DD, which utilizes the dominant difference norm to tackle MaOPs. Experimental results demonstrate that the algorithm is competitive with state-of-the-art approaches on benchmark problems.
IEEE TRANSACTIONS ON CYBERNETICS
(2021)
Article
Management
Deepayan Chakrabarti
Summary: The paper introduces a parameter-free and scalable method called AlphaRob for optimizing the Sharpe ratio, which combines robust optimization and a new concept of portfolio regret. AlphaRob significantly outperforms competing methods on several datasets, showing better performance in optimizing the Sharpe ratio.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2021)
Article
Mathematics, Applied
Krishnendu Adhikary, Samarjit Kar
Summary: In recent years, the concern over increasing greenhouse gas emissions has led to the development of a model for a single-period multi-objective inventory problem that considers uncertainties associated with perishable goods. This model aims to maximize expected profit while minimizing total carbon emissions. To address complexities and uncertainties, credibility theory is used to transform random-fuzzy variables into a crisp equivalent form, enabling more effective analysis and decision-making. The NP-hard nature of the problem is tackled using Multi-objective Particle Swarm Optimization (MOPSO), and the obtained results are validated using a Non-dominated Sorting Genetic Algorithm-II.
COMPUTATIONAL & APPLIED MATHEMATICS
(2023)
Article
Computer Science, Information Systems
Vipin Kumar, Prem Shankar Singh Aydav, Sonajharia Minz
Summary: This paper proposes a multi-view ensemble learning method using multi-objective particle swarm optimization, which finds an optimal solution by balancing the number of views and classification accuracy. Experimental results demonstrate the effectiveness and efficiency of this method on high-dimensional datasets.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Hardware & Architecture
Saba Kanwal, Irfan Younas, Maryam Bashir
Summary: Convolution autoencoders have shown useful applications in image classification and generation. The architecture of deep neural networks greatly impacts network performance, and metaheuristics like multi-objective Particle Swarm Optimization have become popular for optimizing these architectures. The proposed method in this paper enhances the exploration of the search space and results in more generalized, optimized, and accurate architectures for image classification.
COMPUTERS & ELECTRICAL ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Dingming Wu, Xiaolong Wang, Shaocong Wu
Summary: This paper proposes a portfolio construction method based on the continuous trend characteristics of the market to address the challenges of selecting qualified stocks, calculating weights, and considering risk aversion in investment evaluation. By using the k-means clustering algorithm, the method divides stock pools and revises the calculation of returns for the Sharpe ratio, while combining inverse volatility weighting, risk parity, and Markowitz's portfolio theories. Experimental results confirm the superiority of this proposed method in optimizing portfolio construction.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Engineering, Aerospace
Dong-Sun Lee, Hong-Gye Sung
Summary: The performance design of a turbofan engine is optimized using a multi-objective particle swarm optimization method to achieve higher thrust and less fuel consumption. The optimization framework is able to generate well-spread Pareto fronts and reveals the relationship between design variables and engine performance through parallel coordinates visualization.
INTERNATIONAL JOURNAL OF AERONAUTICAL AND SPACE SCIENCES
(2022)
Article
Engineering, Marine
Ahmadreza Eskandari, Ramin Vatankhah, Ehsan Azadi
Summary: This study proposes a nonlinear feedback control strategy for horizontal axis variable speed wind turbines operating below-rated wind speed. The strategy aims to maximize wind energy extraction while reducing mechanical load. The proposed controller compensates for external disturbances, measurement noises, and unmodeled dynamics. It uses sliding mode control combined with backstepping scheme and fuzzy logic system to achieve better performance and prevent excessive mechanical loads on the transmission shaft.
Article
Mathematics
Lesly Lisset Ortiz-Cerezo, Alin Andrei Carsteanu, Julio Bernardo Clempner
Summary: This study proposes a solution for determining the Sharpe ratio portfolio using second order cone programming, and employs a Markov chain structure to represent the asset price process. The efficiency and efficacy of the suggested method are demonstrated using a numerical example.
Article
Computer Science, Artificial Intelligence
Mahdi Dhaini, Nashat Mansour
Summary: Portfolio optimization is a critical financial engineering problem that involves models such as Mean-Variance and Sharpe. Researchers have designed the Squirrel Search Algorithm based on nature-inspired algorithms to address this issue, demonstrating its superiority through comparative analysis and different performance indicators.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Qinghua Gu, Qian Wang, Xuexian Li, Xinhong Li
Summary: A new algorithm, RFMOPSO, is proposed in this paper to optimize constrained combinatorial optimization problems by combining multi-objective particle swarm optimization with a random forest model. Adaptive ranking strategy and novel rule are employed to improve search speed and adaptively update particle states for better objective balance and feasible solutions. Experimental results show promising performance on benchmark problems with discrete variables varying from 10 to 100.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Information Systems
Massimiliano Kaucic, Filippo Piccotto, Gabriele Sbaiz, Giorgio Valentinuz
Summary: This paper proposes a hybrid variant of the level-based learning swarm optimizer (LLSO) to solve large-scale portfolio optimization problems. The algorithm extends the classical mean-variance formulation by maximizing a modified version of the Sharpe ratio while satisfying cardinality, box, and budget constraints. Experimental results show that including this procedure improves solution accuracy.
INFORMATION SCIENCES
(2023)