China
Africa
Explore chapters and articles related to this topic
Introduction
Published in Vlad P. Shmerko, Svetlana N. Yanushkevich, Sergey Edward Lyshevski, Computer Arithmetics for Nanoelectronics, 2018
Vlad P. Shmerko, Svetlana N. Yanushkevich, Sergey Edward Lyshevski
Biologically inspired computing is defined as a paradigm that typifies comprehended biological analogous of computing. Examples of biologically inspired computing are artificial neural networks and evolutionary computing. Artificial neural networks are an attempt to utilize naturalcentric computing (with just moderate success to date) by implementing parallel processing capabilities and networked structures. The methods of contemporary logic design such as data structures design, their optimization, computer array structures, and others typify, to some extent, natural processing.Evolutionary computing centers on search and optimization, which are observed in living systems. Evolutionary algorithms have been used extensively in evolvable hardware [44]. Evolutionary algorithms have been shown to perform well in exploring large and complex design spaces, including logic network design.
The vision of application of multiobjective optimization and genetic algorithm in modeling and simulation of the riser reactor of a fluidized catalytic cracking unit: A critical review
Published in A. K. Haghi, Lionello Pogliani, Eduardo A. Castro, Devrim Balköse, Omari V. Mukbaniani, Chin Hua Chia, Applied Chemistry and Chemical Engineering, 2017
GAs mimic the principles of natural genetics and natural selection to constitute search and optimization procedures. Simulated annealing mimics the cooling phenomenon of molten metals to constitute a search procedure. Professor John Holland of University of Michigan, Ann Arbor, USA envisaged the concept of these GAs in the mid-1960s and published his phenomenal work. Bioinspired optimization mimics traditional biological sciences. Bioinspired, short form for biologically inspired, computing is a field of study that integrates subfields related to the topics of connectionism, social behavior, and emergence. It is often related to the field of artificial intelligence. Biologically inspired computing is a branch of natural computation. Nontraditional optimization involves GA, simulated annealing, and bioinspired optimization.24–26
Bio-Inspired Computing and IoT Networks
Published in Sanjeev J. Wagh, Manisha Sunil Bhende, Anuradha D. Thakare, Energy Optimization Protocol Design for Sensor Networks in IoT Domains, 2023
Sanjeev J. Wagh, Manisha Sunil Bhende, Anuradha D. Thakare
There are many computational methods and tools inspired by nature termed biologically inspired computing. An immune system is an event-response system occurring naturally which quickly adapts and adjusts to changing situations. The characteristics of an Immune System are alert, identify and neutralize the effect of foreign particles in the body. The biological immune system is the complex network which involves tissues, organs, and cell [12]. The function of an immune system is to recognize foreign elements in our body and respond to eliminate or neutralize them. Several researchers adopt and mimic the strategy for developing new techniques for computational intelligence.
Ensemble Classifier for Stock Trading Recommendation
Published in Applied Artificial Intelligence, 2022
Artificial Neural Network (ANN) is biologically inspired computing models which can be used to find knowledge, patterns, or models from a large amount of data (Bose and Liang 1996). ANN consists of connected computational units or nodes, called neurons, arranged in several layers. Each neuron combines its inputs, each of which is multiplied with a weight, and then passes it through an activation function, which can be a linear or nonlinear filter (such as sigmoid or tanh functions). Then the neuron sends its output to other neurons or to be output of the network. The interconnection of all neurons forms different types of architectures designed for various functions. The most widely used architecture is called feed-forward multilayer perceptron (MLP) which generally used back-propagation (BP) learning algorithm to adjust its weights for supervised learning. BP works in an iteration of three steps. The first step is propagating inputs forward through the hidden layers to the output nodes. The second step is to propagate the errors backward through the network starting from output layer. And the final step is to update the weight and biases using approximate steepest descent rule. Such three steps repeat until reaching the maximum iterations allowed with the aim to minimize the errors of output during the training.