Multiobjective Infill Criterion Driven Gaussian Process-Assisted Particle Swarm Optimization of High-Dimensional Expensive Problems

Tian, Jie; Tan, Ying; Zeng, Jianchao; Sun, Chaoli; Jin, Yaochu

doi:10.1109/tevc.2018.2869247

Cited by 147 publications

(56 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are also studies that separately select these two types of samples in the individual-based strategies, for instance in [2], [50]. Most recently, a multi-objective infill criterion has been proposed [56], which considers the infill sampling as a bi-objective problem that simultaneously minimizes the predicted fitness and the estimated variance of the predicted fitness. Then, the solutions on the first and last non-dominated fronts are chosen as new infill samples.…”

Section: A Data Collectionmentioning

confidence: 99%

Data-Driven Evolutionary Optimization: An Overview and Case Studies

Jin

Wang

Chugh

et al. 2019

IEEE Trans. Evol. Computat.

Self Cite

463

144

View full text Add to dashboard Cite

Most evolutionary optimization algorithms assume that the evaluation of the objective and constraint functions is straightforward. In solving many real-world optimization problems, however, such objective functions may not exist. Instead, computationally expensive numerical simulations or costly physical experiments must be performed for fitness evaluations. In more extreme cases, only historical data are available for performing optimization and no new data can be generated during optimization. Solving evolutionary optimization problems driven by data collected in simulations, physical experiments, production processes, or daily life are termed data-driven evolutionary optimization. In this paper, we provide a taxonomy of different data driven evolutionary optimization problems, discuss main challenges in data-driven evolutionary optimization with respect to the nature and amount of data, and the availability of new data during optimization. Real-world application examples are given to illustrate different model management strategies for different categories of data-driven optimization problems.

show abstract

Section: A Data Collectionmentioning

confidence: 99%

Data-Driven Evolutionary Optimization: An Overview and Case Studies

Jin

Wang

Chugh

et al. 2019

IEEE Trans. Evol. Computat.

Self Cite

463

144

View full text Add to dashboard Cite

show abstract

“…That is, by building suitable surrogates based on evaluated data, the DDEAs are able to employ these surrogates to replace the real FEs and then reduce the needs for accessing real FEs. Therefore, DDEAs can have more advantages than traditional EAs when solving expensive and computationally intensive problems [10], [13].…”

Section: A Data-driven Evolutionary Algorithmsmentioning

confidence: 99%

“…In addition, as evaluating promising and uncertain individuals have different advantages, many strategies called infill criteria are proposed and studied based on the combinations of them, such as expected lower confidence bound [19], probability of improvement [59], and expected improvement [52], [60]. Moreover, Tian et al [13] proposed a multiobjective infill criterion driven GP-assisted social learning PSO (MGP-SLPSO), where the multiobjective infill criteria are shown to be efficient when optimizing fitness and minimizing uncertainty together in solving high dimensional problems.…”

Section: B Related Workmentioning

confidence: 99%

Boosting Data-Driven Evolutionary Algorithm With Localized Data Generation

Zhan

Wang

et al. 2020

IEEE Trans. Evol. Computat.

129

View full text Add to dashboard Cite

By efficiently building and exploiting surrogates, data-driven evolutionary algorithms (DDEAs) can be very helpful in solving expensive and computationally intensive problems. However, they still often suffer from two difficulties. First, many existing methods for building a single ad hoc surrogate are suitable for some special problems but may not work well on some other problems. Second, the optimization accuracy of DDEAs deteriorates if available data are not enough for building accurate surrogates, which is common in expensive optimization problems. To this end, this article proposes a novel DDEA with two efficient components. First, a boosting strategy (BS) is proposed for self-aware model managements, which can iteratively build and combine surrogates to obtain suitable surrogate models for different problems. Second, a localized data generation (LDG) method is proposed to generate synthetic data to alleviate data shortage and increase data quantity, which is achieved by approximating fitness through data positions. By integrating the BS and the LDG, the BDDEA-LDG algorithm is able to improve model accuracy and data quantity at the same time automatically according to the problems at hand. Besides, a tradeoff is empirically considered to strike a better balance between the effectiveness of surrogates and the time cost for building them. The experimental results show that the proposed BDDEA-LDG algorithm can generally outperform both traditional methods without surrogates and other state-of-the-art DDEA son widely used benchmarks and an arterial traffic signal timing real-world optimization problem. Furthermore, the proposed BDDEA-LDG algorithm can use only about 2% computational budgets of traditional methods for producing competitive results.

show abstract

“…We expand on each equation of Equation (18), and then subtract the last equation from each formula to get Equation (19). Finally, we use Equations (20)-(23) to simplify the form of the matrix solution.…”

Section: Original Dv-hop Algorithmmentioning

confidence: 99%

“…The PSO algorithm is a classic intelligent algorithm inspired by the foraging behavior of birds. Many scholars have developed and improved on this basis to enhance the global optimization of PSO algorithms [18,19], with developments such as parallel PSO (PPSO) [20], adaptive PSO (APSO) [21], and so on. Differential evolution (DE) [5,5] is a stochastic model that simulates the evolution of organisms.…”

Section: Introductionmentioning

confidence: 99%

Symbiotic Organism Search Algorithm with Multi-Group Quantum-Behavior Communication Scheme Applied in Wireless Sensor Networks

Chu

Pan

2020

Applied Sciences

View full text Add to dashboard Cite

The symbiotic organism search (SOS) algorithm is a promising meta-heuristic evolutionary algorithm. Its excellent quality of global optimization solution has aroused the interest of many researchers. In this work, we not only applied the strategy of multi-group communication and quantum behavior to the SOS algorithm, but also formed a novel global optimization algorithm called the MQSOS algorithm. It has speed and convergence ability and plays a good role in solving practical problems with multiple arguments. We also compared MQSOS with other intelligent algorithms under the CEC2013 large-scale optimization test suite, such as particle swarm optimization (PSO), parallel PSO (PPSO), adaptive PSO (APSO), QUasi-Affine TRansformation Evolutionary (QUATRE), and oppositional SOS (OSOS). The experimental results show that MQSOS algorithm had better performance than the other intelligent algorithms. In addition, we combined and optimized the DV-hop algorithm for node localization in wireless sensor networks, and also improved the DV-hop localization algorithm to achieve higher localization accuracy than some existing algorithms.

show abstract

Multiobjective Infill Criterion Driven Gaussian Process-Assisted Particle Swarm Optimization of High-Dimensional Expensive Problems

Cited by 147 publications

References 58 publications

Data-Driven Evolutionary Optimization: An Overview and Case Studies

Data-Driven Evolutionary Optimization: An Overview and Case Studies

Boosting Data-Driven Evolutionary Algorithm With Localized Data Generation

Symbiotic Organism Search Algorithm with Multi-Group Quantum-Behavior Communication Scheme Applied in Wireless Sensor Networks

Contact Info

Product

Resources

About