A review of techniques for online control of parameters in swarm intelligence and evolutionary computation algorithms

Parpinelli, Rafael Stubs; Plichoski, Guilherme Felippe; Silva, Renan Samuel da; Narloch, Pedro Henrique

doi:10.1504/ijbic.2019.10018955

Cited by 9 publications

(7 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Being gradient‐free and population‐based, these algorithms find optimal and multiple near‐optimal solutions of complex problems that otherwise would be hard to obtain using traditional optimization methods. [ 24,153 ] The involvement of heuristics and populations of candidates can, however, be time‐consuming in obtaining solutions. [ 5 ]…”

Section: Common Ai‐based Process Control Technologiesmentioning

confidence: 99%

“…[ 159 ] Memetic algorithms [ 160 ] and differential evolution algorithm [ 161 ] are variants of GAs that include additional domain‐specific search heuristics. [ 24 ] On the other hand, PSO algorithms move or transform candidates using formulas based on the best‐known positions of individual candidates as well as the swarm or population in the search domain. Similarly, ACO algorithms mimic ants in finding the solution as the optimum cost path in the search domain.…”

Section: Common Ai‐based Process Control Technologiesmentioning

confidence: 99%

“…Such algorithms have been reported to provide superior predictive performances and optimization results in difficult scenarios. [ 24 ] These algorithms and reinforcement learning techniques find optimal solutions from several possibilities. Supervised learning with feedback‐based reinforcement learning forms the basis for Bellman's dynamic programming in machine learning.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Artificial intelligence‐based process control in chemical, biochemical, and biomedical engineering

Dutta

Upreti

2021

Can J Chem Eng

View full text Add to dashboard Cite

In the last three decades, artificial intelligence (AI) has been increasingly and vigorously utilized for process control in chemical, biochemical, and biomedical engineering. These disciplines frequently involve fairly sophisticated processes under risks of operational upsets, and thus have an ever increasing demand of superior control strategies. As the deployment of AI in process control pushes the limits in this regard, the research advances, which are multitudinous and varied, need to be assimilated and organized to help promote further utilization and progress in the field. To that end, we examine more than 280 relevant research publications, and systematically collate the information. The AI‐based technologies are classified, and their over‐arching control paradigm is presented. Common AI‐based control technologies are then presented, which are based on expert systems, fuzzy logic, artificial neural networks, nature‐inspired algorithms, and hybrid approaches. Their working principles, types, and implementations are summarized along with advantages, limitations, and comparisons, if available. Important applications in the above disciplines are included with the help of tables that capture important details. A discussion is also provided on advanced and newly emerging AI‐based control technologies with pertinent applications. Overall trends are analyzed, and future prospects are identified based on the survey.

show abstract

Section: Common Ai‐based Process Control Technologiesmentioning

confidence: 99%

Section: Common Ai‐based Process Control Technologiesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Artificial intelligence‐based process control in chemical, biochemical, and biomedical engineering

Dutta

Upreti

2021

Can J Chem Eng

View full text Add to dashboard Cite

show abstract

“…Meta-heuristic algorithms heavily rely on the values of their parameters, and finding appropriate parameter settings is a complex task that has received a lot of attention. There are two main approaches to parameter tuning: offline (Eryoldaş and Durmuşoglu, 2022) and online (Parpinelli et al, 2019) procedures. Offline tuning (Eryoldaş and Durmuşoglu, 2022) aims to find suitable parameter settings before deploying the algorithm, but traditionally it has been done through trial and error, which is time-consuming, error-prone, and uneven across different algorithms.…”

Section: Introductionmentioning

confidence: 99%

“…However, these methods typically provide a fixed parameter setting for a particular instance. An alternative approach is online tuning (Parpinelli et al, 2019), which involves modifying the parameter settings while solving a problem instance. This approach has the potential advantage of adapting better to the particular instance's characteristics and identifying the best settings for exploratory and exploitative search phases.…”

Section: Introductionmentioning

confidence: 99%

Building Blocks Driven Self-Adaptation in Ant Colony Optimisation

Gaddam,

Barca,

Nguyen

et al. 2023

Preprint

View full text Add to dashboard Cite

This paper presents the implementation and evaluation of the ant colony optimisation with adaptive building blocks, an approach designed to improve the performance of the ant colony optimisation algorithm by selecting self-adaptive building blocks through Elitist, Roulette, and Hybrid (combination between Elitist and Roulette) selection strategies. Ant colony optimisation's performance is heavily dependent on the selection of its parameters, including the building blocks, which determine the search space.The proposed approach aims to enhance the ant colony optimisation algorithm's performance by adapting the building blocks during the optimisation process, with the goal of improving the exploration and exploitation of the search space. The approach selects the most promising building blocks in a dynamic process based on their performance in previous iterations and updates them accordingly. The proposed algorithm was evaluated using the CEC'2013 benchmark functions, which are commonly used to evaluate optimisation algorithms. The results of the experiments demonstrate that Elitist selection outperformed all other algorithms, including the standard ant colony optimisation, Roulette, and Hybrid. Elitist selection of building blocks showed the best performance in terms of computational time and the number of iterations required to reach the optimal fitness value.

show abstract

A Self-adaptive Differential Evolution with Fragment Insertion for the Protein Structure Prediction Problem

Silva

Parpinelli

2018

Hybrid Metaheuristics

View full text Add to dashboard Cite

A review of techniques for online control of parameters in swarm intelligence and evolutionary computation algorithms

Cited by 9 publications

References 0 publications

Artificial intelligence‐based process control in chemical, biochemical, and biomedical engineering

Artificial intelligence‐based process control in chemical, biochemical, and biomedical engineering

Building Blocks Driven Self-Adaptation in Ant Colony Optimisation

A Self-adaptive Differential Evolution with Fragment Insertion for the Protein Structure Prediction Problem

Contact Info

Product

Resources

About