Article
Computer Science, Artificial Intelligence
Adriana Menchaca-Mendez, Saul Zapotecas-Martinez, Luis Miguel Garcia-Velazquez, Carlos A. Coello Coello
Summary: Design of experiments is a statistical branch widely applied in various fields. This paper proposes a new method to approximate uniform mixture designs using evolutionary multi-objective optimization, and demonstrates its effectiveness through experiments.
SWARM AND EVOLUTIONARY COMPUTATION
(2022)
Article
Energy & Fuels
Wenpeng Luan, Longfei Tian, Bochao Zhao
Summary: Dynamic tariffs are essential in demand response as they help smooth power consumption, reduce generation capacity requirement, and carbon emissions. However, existing works often overlook important factors such as user responses to tariffs when designing them. To address this issue, this paper proposes a new dynamic tariff design method that considers user responses to tariff changes. The method utilizes non-intrusive load monitoring technique to acquire information on rated power and user preferences for each appliance, which is then used to quantify user comfort or discomfort based on their appliance usage habits. A bi-level Stackelberg game model is then built to design optimal dynamic tariffs and simulate the impact of tariff changes on users' demand response plans. The results show that the proposed model generally outperforms benchmark methods in achieving peak shaving, low carbon emission, and user satisfaction.
Article
Engineering, Aerospace
Pau Garcia Buzzi, Daniel Selva
Summary: With advancements in satellite technology and cost reduction, Distributed Spacecraft Missions (DSMs) have emerged as a new paradigm in space exploration, offering higher capabilities through the use of multiple simpler and affordable satellites. However, the orbit selection problem for Earth Observation DSMs has become more complex, requiring consideration of multiple design variables and conflicting objectives.
Article
Computer Science, Artificial Intelligence
Ke-Jing Du, Jian-Yu Li, Hua Wang, Jun Zhang
Summary: Evolutionary multi-objective multi-task optimization is an emerging paradigm for solving multi-objective multi-task optimization problems using evolutionary computation. This paper proposes treating these problems as multi-objective multi-criteria optimization problems and develops an algorithm framework that utilizes the knowledge of all tasks in the same population. The algorithm selects fitness evaluation functions as criteria, guided by a probability-based selection strategy and an adaptive parameter learning method. Extensive experiments show the effectiveness and efficiency of the proposed algorithm. Treating MO-MTOP as MO-MCOP is a potential and promising direction for solving these problems.
COMPLEX & INTELLIGENT SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Bo Jiang, Hongtao Lei, Wenhua Li, Rui Wang
Summary: This paper investigates the design of hybrid renewable energy systems (HRES) and proposes a novel multi-objective evolutionary algorithm with a special environmental selection strategy to enhance the diversity of solutions. The effectiveness, superiority, and generalizability are validated through experiments and comparison with state-of-the-art algorithms.
SWARM AND EVOLUTIONARY COMPUTATION
(2022)
Article
Automation & Control Systems
Yuhong Li, Ni Li, Guanghong Gong, Jin Yan
Summary: The paper proposes an intelligent design of experiment algorithm using an improved evolutionary multi-objective optimization approach, which shows better sampling capacity and fine sampling efficiency compared to existing algorithms. The application effects of a complex flight simulator demonstrate the algorithm's wide technological prospect in serving certain complex systems well.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2021)
Article
Computer Science, Artificial Intelligence
Elaine Guerrero-Pena, Aluizio F. R. Araujo
Summary: Dynamic multi-objective evolutionary algorithms can address multi-objective optimization problems by predicting and responding to changes, with prediction-based methods showing promise. Through the use of objective space prediction strategy and change reaction mechanism, the proposed DOSP-NSDE demonstrates competitiveness in experiments.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Artificial Intelligence
David Judt, Craig Lawson, Albert S. J. van Heerden
Summary: As aircraft systems design becomes more integrated with aircraft structural definition, flexible FQI design methods are needed for assessing system-level impact due to aircraft level changes. The proposed FQI-GA method, with two-stage fitness assignment and FQI specific crossover procedure, can handle multiple measurement accuracy constraints and is suitable for assessing aircraft fuel quantity indication system design. Results from testing demonstrate the effectiveness of the method in quickly investigating FQI probe layouts and trade-offs.
INTEGRATED COMPUTER-AIDED ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Yongkuan Yang, Jianchang Liu, Shubin Tan
Summary: Many MOEAs are developed to solve CMOPs, but they encounter low efficiency for steady-state CMOPs. This paper proposes a multi-objective evolutionary algorithm named FACE, which maintains the known feasible solution in the second population and evolves together with the main population. Performance comparisons show the efficiency and scalability of FACE for steady-state CMOPs.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Information Systems
Ruochen Liu, Ping Yang, Haoyuan Lv, Weibin Li
Summary: This article proposes a multi-factorial evolutionary algorithm (MFEA) for solving the container placement problem in heterogeneous cluster environments. The MFEA algorithm, embedded with a local search strategy, significantly reduces optimization time and provides competitive solutions for container placement.
IEEE TRANSACTIONS ON CLOUD COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Edgar Galvan, Fergal Stapleton
Summary: This study makes progress in neuroevolution for vehicle trajectory prediction by adopting rich artificial neural networks and two evolutionary multi-objective optimization algorithms. The underlying mechanisms and response to objective scaling of each algorithm are revealed. Additionally, certain objectives are found to be beneficial while others are detrimental to finding valid models.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Theory & Methods
Shouyong Jiang, Juan Zou, Shengxiang Yang, Xin Yao
Summary: Evolutionary dynamic multi-objective optimisation (EDMO) is a rapidly growing area that uses evolutionary approaches to solve multi-objective optimisation problems with time-varying changes. After nearly two decades, significant advancements have been made in theoretic research and applications. This article provides a comprehensive survey and taxonomy of existing research on EDMO, as well as highlighting multiple research opportunities for further development.
ACM COMPUTING SURVEYS
(2023)
Article
Chemistry, Medicinal
Flavia Varano, Daniela Catarzi, Erica Vigiani, Diego Dal Ben, Michela Buccioni, Gabriella Marucci, Lorenzo Di Cesare Mannelli, Elena Lucarini, Carla Ghelardini, Rosaria Volpini, Vittoria Colotta
Summary: New compounds with high affinity and antidepressant-like activity were synthesized and evaluated in vitro for their potential as A(1) and A(2A) adenosine receptor ligands.
Article
Computer Science, Hardware & Architecture
Wenyi Zheng, Abolfazl Mehbodniya, Rahul Neware, Surindar Gopalrao Wawale, Bibhu Prasad Ganthia, Mohammad Shabaz
Summary: This study aims to design and develop a modular unmanned aerial vehicle (UAV) platform for high-precision collection and real-time integration of air quality data. The UAV can fly on predetermined pathways to collect data within adequate flight time.
COMPUTERS & ELECTRICAL ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Maysam Orouskhani, Daming Shi, Yasin Orouskhani
Summary: This paper introduces a novel multi-objective evolutionary clustering algorithm based on centrality modularity, which uses node similarity to determine the optimal initial population and structural modularity to automatically determine the optimal number of clusters. Experimental results show that the proposed algorithm outperforms traditional methods in terms of performance.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Dani Irawan, Boris Naujoks, Thomas Back, Michael Emmerich
Summary: This study proposes an improved control variable analysis based on dominance and diversity in Pareto optimization, and applies it in a cooperative coevolution framework with orthogonal sampling mutation. The results demonstrate that the proposed method outperforms the traditional method in terms of accuracy and competitiveness.
Article
Medicine, Legal
Peter S. R. Wright, Katharine A. Briggs, Robert Thomas, Graham F. Smith, Gareth Maglennon, Paulius Mikulskis, Melissa Chapman, Nigel Greene, Benjamin U. Phillips, Andreas Bender
Summary: By comparing histopathological findings and target organ toxicities across different preclinical species, this study found that positive concordance is more common than negative concordance in histopathological results, and there is low concordance in target organ toxicities. It provides new statistically significant associations between preclinical species but finds that concordance is rare.
REGULATORY TOXICOLOGY AND PHARMACOLOGY
(2023)
Article
Pharmacology & Pharmacy
Hongbin Yang, Olga Obrezanova, Amy Pointon, Will Stebbeds, Jo Francis, Kylie A. Beattie, Peter Clements, James S. Harvey, Graham F. Smith, Andreas Bender
Summary: Functional changes to cardiomyocytes during drug discovery pose risks of cardiovascular adverse effects. A new approach using calcium transients in hiPSC-CMs has been developed to detect early contractility changes. By deriving 25 parameters from each calcium transient waveform, a modified Random Forest method was able to predict inotropic effects with improved accuracy compared to traditional methods. This study demonstrates the potential of advanced waveform parameters and machine learning techniques in predicting cardiovascular risks associated with inotropic effects.
TOXICOLOGY AND APPLIED PHARMACOLOGY
(2023)
Article
Pharmacology & Pharmacy
Yingli Zhu, Hongbin Yang, Liwen Han, Lewis H. Mervin, Layla Hosseini-Gerami, Peihai Li, Peter Wright, Maria-Anna Trapotsi, Kechun Liu, Tai-Ping Fan, Andreas Bender
Summary: Uncontrolled angiogenesis is a common problem in many deadly and debilitating diseases, and traditional Chinese medicine offers an alternative source for developing drugs to regulate angiogenesis. In this study, 100 traditional Chinese medicine-derived metabolites were investigated, and 51 metabolites were found to have angiogenic activity. The mechanisms of action of these metabolites were analyzed, and a decision tree was generated to predict their poly-pharmacology. In vitro and in vivo experiments were conducted to validate the predictions and identify specific metabolites with pro-angiogenic or anti-angiogenic effects.
FRONTIERS IN PHARMACOLOGY
(2023)
Review
Biochemical Research Methods
Anika Liu, Srijit Seal, Hongbin Yang, Andreas Bender
Summary: This review discusses various sources of information, including biological data such as gene expression and cell morphology, for better understanding and predicting compound activity and safety-related endpoints. It introduces different types of chemical, in vitro, and in vivo information that can describe compounds and adverse effects. The review explores how compound descriptors based on chemical structure or biological perturbation response can predict safety-related endpoints, and how biological data can enhance understanding of adverse effects mechanistically. These applications highlight the potential of large-scale biological information in predictive toxicology and drug discovery projects.
Article
Biochemical Research Methods
Layla Hosseini-Gerami, Ixavier Alonzo Higgins, David A. Collier, Emma Laing, David Evans, Howard Broughton, Andreas Bender
Summary: This study performed a comprehensive evaluation of four causal reasoning algorithms in different networks, and found that the choice of algorithm and network greatly influenced the performance of causal reasoning algorithms. SigNet performed best in recovering direct targets, while CARNIVAL with Omnipath network excelled in recovering informative signaling pathways. The performance of causal reasoning methods was somewhat correlated with the connectivity and biological role of the targets.
BMC BIOINFORMATICS
(2023)
Article
Biochemistry & Molecular Biology
Sohvi Luukkonen, Helle W. van den Maagdenberg, Michael T. M. Emmerich, Gerard J. P. van Westen
Summary: The factors determining a drug's success are diverse, making drug design a multi-objective optimisation problem. With the emergence of machine learning and optimisation methods, there has been a rapid increase in developments and applications in the field of multi-objective compound design. Population-based metaheuristics and deep reinforcement learning are commonly used methods, but conditional learning methods are gaining popularity. This article provides a brief overview of the field and the latest innovations.
CURRENT OPINION IN STRUCTURAL BIOLOGY
(2023)
Article
Biochemistry & Molecular Biology
Benoit Baillif, Jason Cole, Patrick McCabe, Andreas Bender
Summary: Deep generative models have become popular in chemical design. This article focuses on explicit 3D molecular generative models, which have gained interest recently. Multiple models have been developed to generate molecules in 3D, providing atom types and coordinates. These models can be guided by structural information and produce molecules with similar docking scores to known actives, but they are less efficient and sometimes generate unrealistic conformations. The article advocates for a unified benchmark of metrics and proposes future perspectives to be addressed.
CURRENT OPINION IN STRUCTURAL BIOLOGY
(2023)
Article
Chemistry, Medicinal
Brandon J. Bongers, Huub J. Sijben, Peter B. R. Hartog, Andrey Tarnovskiy, Adriaan P. IJzerman, Laura H. Heitman, Gerard J. P. van Westen
Summary: In this study, a computational screening pipeline was developed to find new inhibitors for the NET protein. A data-driven approach was used to diversify the chemical space and select optimal proteins to model for NETs. A proteochemometric model was created and applied to an extensive compound database, resulting in the identification of five potential hit compounds with promising inhibitory potencies toward NET.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2023)
Article
Medicine, Research & Experimental
Koichi Handa, Peter Wright, Saki Yoshimura, Michiharu Kageyama, Takeshi Iijima, Andreas Bender
Summary: This study developed machine learning models to predict plasma concentration-time profiles of drugs after intravenous and oral administration. The predictive accuracy of different models was investigated, and random forest showed the best performance. The importance of in vitro pharmacokinetic parameters was also explored and found to be well-reflected in the model.
MOLECULAR PHARMACEUTICS
(2023)
Article
Computer Science, Interdisciplinary Applications
Kegong Diao, Michael Emmerich, Jacob Lan, Iryna Yevseyeva, Robert Sitzenfrei
Summary: This paper introduces a multi-objective optimization approach for efficient sensor placement in water distribution networks. By minimizing the expected detection time and maximizing the detection network coverage, the minimal number of required sensors can be identified. The approach is tested on a benchmark problem with 129 nodes, and the results demonstrate the significant improvement in detection network coverage by deploying only a few sensors. Additionally, the study shows that 40-45 sensors are sufficient for fully monitoring the benchmark network.
JOURNAL OF HYDROINFORMATICS
(2023)
Article
Environmental Sciences
Juhuhn Kim, Michael T. M. Emmerich, Robert Voors, Barend Ording, Jong-Seok Lee
Summary: This article proposes an unsupervised clustering technique for monitoring shipping emissions based on NO2 concentration. The method is tested and validated using data from multiple regions and time periods, improving the spatial correlation coefficients between NO2 column clusters and shipping traffic frequency. Additionally, a temporal correlation is identified between NO2 column levels along shipping routes and the global container throughput index.
Article
Chemistry, Medicinal
Lavinia-Lorena Pruteanu, Andreas Bender
Summary: Gene expression and cell morphology data are valuable in drug discovery, providing insight into biological systems in different states and after compound treatment. This article discusses recent advances in using these data for drug repurposing and highlights the need for further understanding of the applicability domain and relevance of the readouts for decision making.
ACS MEDICINAL CHEMISTRY LETTERS
(2023)
Article
Chemistry, Medicinal
Brandon J. Bongers, Huub J. Sijben, Peter B. R. Hartog, Andrey Tarnovskiy, Adriaan P. IJzerman, Laura H. Heitman, Gerard J. P. van Westen
Summary: This study developed a computational screening pipeline to identify new inhibitors for the high-affinity norepinephrine transporter (NET). By using the chemical space of related proteins, a data-driven approach was used to diversify the known chemical space for NET modeling. The final model, created through a two-step approach, predicted 46 chemically diverse candidates, of which five compounds showed promising inhibitory potency towards NET in experimental assays.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2023)
Article
Mathematics, Interdisciplinary Applications
Hao Wang, Michael Emmerich, Andre Deutz, Victor Adrian Sosa Hernandez, Oliver Schutze
Summary: The Hypervolume Newton Method (HVN) is proposed as a fast and precise indicator-based method for solving unconstrained bi-objective optimization problems. It seeks to maximize the hypervolume indicator by adopting the Newton-Raphson method. The HVN method was hybridized with a multi-objective evolutionary algorithm to extend its scope to non-convex optimization problems.
MATHEMATICAL AND COMPUTATIONAL APPLICATIONS
(2023)