Experimental Analysis Between Exploration and Exploitation

Cuevas, Erik; Diaz, Parris; Camarena, Octavio

doi:10.1007/978-3-030-58100-8_10

Cited by 5 publications

(5 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The management of the exploration and exploitation phases by the REE function is too simplistic because it is based on a linear decreasing function. However, poor management between exploration and exploitation and a poor balance between these two phases is the assurance of being stuck in a local optimum and leading to poor final solutions [61,62]. As for exploitation, although the four HHO exploitation equations are of good mathematic quality, they do not systematically focus on promising areas in the case of high-dimensional and highly multimodal optimization problems.…”

Section: Harris Hawk Optimization Encirclement-attack-synergy Overviewmentioning

confidence: 99%

BHHO-EAS metaheuristic applied to the NP-Hard wrapper feature selection multi-objective optimization problem

SASSI,

CHELOUAH

2024

Preprint

View full text Add to dashboard Cite

Faced with the increase in high-dimensional Big Data creating more volume and complexity, the feature selection process became an essential phase in the preprocessing workflow upstream of the design of systems based on deep learning. This paper is a concrete and first application of the new metaheuristic Harris Hawk Optimization Encirclement-Attack-Synergy (HHO-EAS) in solving the NP-Hard wrapper feature selection multi-objective optimization problem. This problem combines two contradictory objectives: maximizing the accuracy of a classifier while minimizing the number of the most relevant and non-redundant selected features. To do this we hybridized HHO-EAS to create the new metaheuristic Binary HHO-EAS (BHHO-EAS). We combined HHO-EAS to the sixteen transfer functions most used in the literature structured in a balanced way among the four main categories including S-Shaped, V-Shaped, Q-Shaped and U-Shaped. This wide range of transfer function allows us to analyze the evolution of BHHO-EAS’s skills according to the assigned transfer function and to determine which of them offer the best performances. We applied wrapper feature selection to the well-known NSL-KDD dataset with the deep learning Multi Layer Perceptron (MLP) classifier. We put BHHO-EAS in competition with three other well-known population based binary metaheuristics, BPSO, BBA and BHHO. The analysis of the experimental results, compared to the three other binary metaheuristics, demonstrated that BHHO-EAS obtained the best performance on 100% of the transfer functions. This is more particularly highlighted by the U-Shaped transfer functions, which give an acceptable compromise for the two objectives of the problem with an average accuracy of 96,4% and an average size of selected features of 20.

show abstract

Section: Harris Hawk Optimization Encirclement-attack-synergy Overviewmentioning

confidence: 99%

BHHO-EAS metaheuristic applied to the NP-Hard wrapper feature selection multi-objective optimization problem

SASSI,

CHELOUAH

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…This phase aims to diversify the set of solutions under consideration, thereby increasing the chances of discovering novel and potentially superior solutions (Hussain et al, 2019). By venturing into unexplored territories, metaheuristics harness a degree of randomness and adaptability, allowing them to escape local optima -suboptimal solutions that are locally optimal but not globally so (Cuevas et al, 2021).…”

Section: Exploration and Exploitationmentioning

confidence: 99%

Metaheuristic Optimization Algorithms: an overview

Dr. Benaissa Brahim,

Kobayashi,

Al Ali

et al. 2024

HCMCOUJS - Advances in Computational Structures

View full text Add to dashboard Cite

Metaheuristic optimization algorithms are known for their versatility and adaptability, making them effective tools for solving a wide range of complex optimization problems. They don't rely on specific problem types, gradients, and can explore globally while handling multi-objective optimization. They strike a balance between exploration and exploitation, contributing to advancements in optimization. However, it's important to note their limitations, including the lack of a guaranteed global optimum, varying convergence rates, and their somewhat opaque functioning. In contrast, metaphor-based optimization algorithms, while intuitively appealing, have faced controversy due to potential oversimplification and unrealistic expectations. Despite these considerations, metaheuristic algorithms continue to be widely used for tackling complex problems. This research paper aims to explore the fundamental components and concepts that underlie optimization algorithms, focusing on the use of search references and the delicate balance between exploration and exploitation. Visual representations of the search behavior of selected metaheuristic algorithms will also be provided.

show abstract

“…It is not within this paper's scope to discuss the correspondence between the operator and its operation. Please refer Črepinšek, Liu, and Mernik (2013); Cuevas, Diaz, and Camarena (2021); Eiben and Schippers (1998) for details.…”

Section: Framework For Sieasmentioning

confidence: 99%

Socio-inspired evolutionary algorithms: a unified framework and survey

2023

View full text Add to dashboard Cite

Evolutionary computation has gone through vast and diverse research endeavors in the past few decades. Although the initial inspiration came from Darwin's ideas of biological evolution, the field has moved thereon to ideas from the collective intelligence of insects, birds, and fish, to name a few. A variety of algorithms have been proposed in the literature based on these ideas and have been shown to perform well in various applications. More recently, inspiration from human behavior and knowledge exchange and transformation has given rise to a new evolutionary computing paradigm. It is well recognized that human societies and problem-solving capabilities have evolved much faster than biological evolution. Many research endeavors have been reported in the literature inspired by diverse aspects of human societies with corresponding terminologies to describe the algorithm. These endeavors have resulted in a plethora of algorithms worded differently from each other, but the underlying mechanisms could be more or less similar, causing immense confusion for a new reader. This paper presents a generalized framework for these Socio-inspired Evolutionary Algorithms (SIEAs) or Socio-inspired Metaheuristic Algorithms. A survey of various SIEAs is provided to highlight the working of these algorithms on a common framework, their variations and improved versions proposed in the literature, and their applications in various fields of search and optimization. The algorithmic description of each SIEA enables a clearer understanding of the similarities and differences between these methodologies. Efforts have been made to provide an extensive list of references with due context. In that sense, this paper could become an excellent reference as a starting point for anyone interested in this fascinating field of research.

show abstract

Experimental Analysis Between Exploration and Exploitation

Cited by 5 publications

References 35 publications

BHHO-EAS metaheuristic applied to the NP-Hard wrapper feature selection multi-objective optimization problem

BHHO-EAS metaheuristic applied to the NP-Hard wrapper feature selection multi-objective optimization problem

Metaheuristic Optimization Algorithms: an overview

Socio-inspired evolutionary algorithms: a unified framework and survey

Contact Info

Product

Resources

About