Mitigating Metaphors: A Comprehensible Guide to Recent Nature-Inspired Algorithms

Lones, Michael A.

doi:10.1007/s42979-019-0050-8

Cited by 64 publications

(36 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the aim of this work is not just to design better optimisers, but also to explore and understand the optimiser design space in a more systematic and context-sensitive manner. This is motivated by recent growth in the design of natureinspired optimisation algorithms, which has seen the invention of new optimisers based on principles of animal foraging and other natural phenomena that are only tangentially related to optimisation [Sörensen, 2015, Lones, 2019. This work uses Push [Spector, 2001, Spector and Robinson, 2002, Spector et al, 2004, a language that was designed to address the need for both expressiveness and evolvability when optimising programs using evolutionary algorithms.…”

Section: Introductionmentioning

confidence: 99%

Instruction-level design of local optimisers using push GP

Lones¹

2019

Proceedings of the Genetic and Evolutionary Computation Conference Companion

Self Cite

View full text Add to dashboard Cite

This work uses genetic programming to explore the design space of local optimisation algorithms. Optimisers are expressed in the Push programming language, a stack-based language with a wide range of typed primitive instructions. The evolutionary framework provides the evolving optimisers with an outer loop and information about whether a solution has improved, but otherwise they are relatively unconstrained in how they explore optimisation landscapes. To test the utility of this approach, optimisers were evolved on four different types of continuous landscape, and the search behaviours of the evolved optimisers analysed. By making use of mathematical functions such as tangents and logarithms to explore different neighbourhoods, and also by learning features of the landscapes, it was observed that the evolved optimisers were often able to reach the optima using relatively short paths.

show abstract

Section: Introductionmentioning

confidence: 99%

Instruction-level design of local optimisers using push GP

Lones¹

2019

Proceedings of the Genetic and Evolutionary Computation Conference Companion

Self Cite

View full text Add to dashboard Cite

show abstract

“…begin [2]. define algorithm list = #"Jaya", "SootyTern", "Owl", "Butterfly", "HenryGas"# [3].…”

Section: Hhgso Implementationmentioning

confidence: 99%

Hybrid Henry gas solubility optimization algorithm with dynamic cluster-to-algorithm mapping

Zamli

Kader

Azad

et al. 2021

Neural Comput & Applic

View full text Add to dashboard Cite

Henry Gas Solubility Optimization Algorithm (HGSO) is a recently developed population-based metaheuristic algorithm in the literature. One notable feature of HGSO is that the algorithm divides its (single) population into a set of clusters that are individually mapped to an independent HGSO with its parameter settings (as well as its local best). At a glance, having multiple independent HGSO serving the given clusters in the population can definitely boost exploration (i.e., in terms of roaming the new potential region in the search space for better solution alternatives). However, a closer look reveals two main limitations. Firstly, HGSO-to-cluster mapping is statically defined. To be specific, the defined HGSO-to-cluster mapping does not consider its adaptive performance for the subsequent iteration. Secondly, HGSO implementation ignores the opportunity for hybridization with other meta-heuristic algorithms. With hybridization, one can compensate the limitation of a host algorithm with other algorithms' strength. Best results in the literature have often been associated with hybridization. Addressing these limitations, this paper proposes the development of Hybrid HGSO (HHGSO). Taking HGSO as the host algorithm, HHGSO is hybridized with four recently developed meta-heuristic algorithms, including Jaya Algorithm (JA), Sooty Tern Optimization Algorithm (STOA), Butterfly Optimization Algorithm (BOA) and Owl Search Algorithm (OSA). The individual mapping of each algorithm is made dynamic based on penalized and reward adaptive probability. Comparative performance of HHGSO with the aforementioned algorithms is conducted with a well-known Searchbased Software Engineering (SBSE) problem involving team formation problem. Additionally, the defined hybridization approach has also been adopted as a hybridization template for solving the combinatorial test generation problem with the same meta-heuristic algorithm combinations. Comparative performance is also undertaken against recently developed hyper-heuristic algorithms involving Exponential Monte Carlo with Counter, Modified Choice Function, Improvement Selection Rules, and Fuzzy Inference Selection. Our results indicate that the HHGSO hybridization has usefully improved the performance of the original HGSO and gives superior performance against the given competing algorithms.

show abstract

“…In the following we try to distinguish classification schemes referring to algorithms from those referring to frameworks. This is also important when considering the novelty of metaheuristics, as new algorithms are easier to design than new frameworks (Lones 2019).…”

Section: Definitionsmentioning

confidence: 99%

“…Algorithm: The fourth level of the classification system uses basic metaheuristic components incorporated in a metaheuristic framework as criteria for classification. A basis for these criteria was established by Lones (2014Lones ( , 2019. His work analysed metaheuristics in order to find basic algorithmic structures incorporated in them.…”

Section: Classificationmentioning

confidence: 99%

Classifying Metaheuristics: Towards a unified multi-level classification system

Stegherr¹,

Heider²,

Hähner³

2020

Nat Comput

View full text Add to dashboard Cite

Metaheuristics provide the means to approximately solve complex optimisation problems when exact optimisers cannot be utilised. This led to an explosion in the number of novel metaheuristics, most of them metaphor-based, using nature as a source of inspiration. Thus, keeping track of their capabilities and innovative components is an increasingly difficult task. This can be resolved by an exhaustive classification system. Trying to classify metaheuristics is common in research, but no consensus on a classification system and the necessary criteria has been established so far. Furthermore, a proposed classification system can not be deemed complete if inherently different metaheuristics are assigned to the same class by the system. In this paper we provide the basis for a new comprehensive classification system for metaheuristics. We first summarise and discuss previous classification attempts and the utilised criteria. Then we present a multi-level architecture and suitable criteria for the task of classifying metaheuristics. A classification system of this kind can solve three main problems when applied to metaheuristics: organise the huge set of existing metaheuristics, clarify the innovation in novel metaheuristics and identify metaheuristics suitable to solve specific optimisation tasks.

show abstract

Mitigating Metaphors: A Comprehensible Guide to Recent Nature-Inspired Algorithms

Cited by 64 publications

References 62 publications

Instruction-level design of local optimisers using push GP

Instruction-level design of local optimisers using push GP

Hybrid Henry gas solubility optimization algorithm with dynamic cluster-to-algorithm mapping

Classifying Metaheuristics: Towards a unified multi-level classification system

Contact Info

Product

Resources

About