Adobe PDF (6.4 MB)
Table of Contents - Adobe PDF (728.68 kB)
Brochure
Download
User comments
Title Details:
Computational intelligence and deep learning
Authors: Likothanassis, Spiridon
Koutsomitropoulos, Dimitrios
Subject: MATHEMATICS AND COMPUTER SCIENCE
MATHEMATICS AND COMPUTER SCIENCE > COMPUTER SCIENCE > INFORMATION MANAGEMENT
MATHEMATICS AND COMPUTER SCIENCE > COMPUTER SCIENCE > INTELLIGENT SYSTEMS
MATHEMATICS AND COMPUTER SCIENCE > COMPUTER SCIENCE > INTELLIGENT SYSTEMS > ADVANCED REPRESENTATION AND REASONING
MATHEMATICS AND COMPUTER SCIENCE > COMPUTER SCIENCE > INTELLIGENT SYSTEMS > ADVANCED MACHINE LEARNING
Keywords:
Neural networks
Deep learning
Genetic algorithms
Convolutional networks
Backpropagation
Genetic programming
Description:
Abstract:
The objective of this book is to be a basic educational material in “Computational Intelligence and Deep Learning”. At first the basic concepts of Artificial Neural Networks and Genetic Algorithms are presented. The association with Artificial Intelligence and the traditional search and optimization methods is discussed, as well as with the biological systems that they have inspired. The basic concepts of learning theory and the two learning paradigms (supervised and unsupervised learning) are provided. It follows the presentation of the basic training algorithms of the Artificial Neural Networks, with focus on the well-known algorithm, Error Back Propagation – EBP). The Deep Learning Networks, models and training, conclude the presentation of feedforward artificial neural networks. Next the Genetic/Evolutionary algorithms are founded and a case study is discussed - their combination in a hybrid algorithm-, and how to design and train an evolutionary neural network as well. Also, it is presented a short introduction to Genetic Programming (GP), that is a permutation of Evolutionary Algorithms, based on the Darwin’s evolution theory. Finally, two different unsupervised learning paradigms are presented. More specifically, the Hopfield Networks (a kind of autocorrelated memory) and Kohonen Networks (a kind of self-organized maps for data clustering) are presented, as well as their training algorithms. In all the chapters the objectives and the learning results are referred and it is quoted a sufficient number of examples, exercises and activities, that help in better understanding of the presented subjects. It will be very helpful for the readers to have basic knowledge of discrete mathematics, linear algebra, combinatorics and programming. Those who are interested to use these technologies for problem solving have to take care to access these skills. Otherwise, they will have troubles with the easy understanding of the material.
Linguistic Editors: Kanari, Vasiliki
Graphic Editors: Moustani, Evagelia
Type: Undergraduate textbook
Creation Date: 01-03-2023
Item Details:
ISBN 978-618-5726-47-8
License: Attribution - NonCommercial - ShareAlike 4.0 International (CC BY-NC-SA 4.0)
DOI http://dx.doi.org/10.57713/kallipos-168
Handle http://hdl.handle.net/11419/9117
Bibliographic Reference: Likothanassis, S., & Koutsomitropoulos, D. (2023). Computational intelligence and deep learning [Undergraduate textbook]. Kallipos, Open Academic Editions. https://dx.doi.org/10.57713/kallipos-168
Language: Greek
Consists of:
1. Capabilities and Applications οf Neural Networks and Genetic Algorithms
2. Introduction to Artificial Neural Networks
3. Learning Algorithms
4. Principles and Restrictions in Designing Artificial Neural Networks
5. Hopfield Nets and Kohonen Nets
6. Optimizations for Deep Model Training
7. Deep Neural Network Models
8. Introduction to Genetic Algorithms
9. Case Study - Evolutionary Neural Networks
10. Genetic Programming (GP)
Number of pages 262
Publication Origin: Kallipos, Open Academic Editions
User comments
There are no published comments available!