NIANN: Integration of ANN with Nature-Inspired Optimization Algorithms
Pati SK, Banerjee A, Gupta MK, Shai R (2023)
Publication Type: Book chapter / Article in edited volumes
Publication year: 2023
Journal
Publisher: Springer Science and Business Media Deutschland GmbH
Edited Volumes: Nature-Inspired Optimization Methodologies in Biomedical and Healthcare
Series: Intelligent Systems Reference Library
Book Volume: 233
Pages Range: 113-145
DOI: 10.1007/978-3-031-17544-2_6
Abstract
Artificial neural networks (ANNs) are stimulated according to the biological brain's connection of axons and dendrons. These neural networks perform a major part in the advancement of artificial intelligence and learning algorithms. Though initially used for image classification, in modern times applications of ANNs have been useful over numerous fields such as medical data mining, bioinformatics, natural language processing, time series forecasting, and in various optimization problems as well. Nature-inspired algorithms are a set of novel problem-solving approaches that are derived from various incidents occurring in nature around us. Each of the methods such as the BAT, genetic algorithm, or colony optimization methods were created by keeping a specific hard problem in mind. In recent times general purpose use of these nature-inspired algorithms has become widely popular in solving mainly optimization problems derived from the fields of NLP, machine learning, deep learning, classification, and feature selection as well. Nature-inspired algorithms mainly work by mimicking phenomena occurring in nature among various species on a macro scale. A set of nature-inspired algorithms such as the genetic algorithm family mimics the processes that occur in a microorganism such as a cell division, mutation, etc. Since these algorithms are inspired by nature and were developed keeping in mind achieving an optimal solution of a given hard problem, their application in general-purpose problems also yields satisfactory results. If an algorithm fails to achieve a satisfactory solution to a problem, it is easy to modify them according to the need of the given problem to overcome any obstacle. In this chapter, an approach is introduced that aims to combine the nature-inspired optimization algorithm with the learning model of artificial neural networks to provide a more accurate and streamlined output generation of the neural network. Nature-inspired algorithms can be used as a learning method in the ANN model. In contrast to that, an ANN can also be used as an objective function to a nature-inspired algorithm to improve its capability to generate an optimal solution. This chapter aims to explore both approaches in detail.
Authors with CRIS profile
Rinita Shai
Graduiertenkolleg 2339/2 IntComSin: Grenzflächen, komplexe Strukturen und singuläre Grenzwerte in der Kontinuumsmechanik - Analysis und Numerik
Involved external institutions
Jalpaiguri Government Engineering College
India (IN)
Maulana Abul Kalam Azad University of Technology (MAKAUT) / West Bengal University of Technology (WBUT)
India (IN)
India (IN)
Maulana Abul Kalam Azad University of Technology (MAKAUT) / West Bengal University of Technology (WBUT)
India (IN)
How to cite
APA:
Pati, S.K., Banerjee, A., Gupta, M.K., & Shai, R. (2023). NIANN: Integration of ANN with Nature-Inspired Optimization Algorithms. In Nature-Inspired Optimization Methodologies in Biomedical and Healthcare. (pp. 113-145). Springer Science and Business Media Deutschland GmbH.
MLA:
Pati, Soumen Kumar, et al. "NIANN: Integration of ANN with Nature-Inspired Optimization Algorithms." Nature-Inspired Optimization Methodologies in Biomedical and Healthcare. Springer Science and Business Media Deutschland GmbH, 2023. 113-145.
BibTeX: Download