About the Book
A Comprehensive Guide to Theory and Implementation.
Bridging the gap between theory and practice with this extensive guide to neural networks, featuring parallel implementations in both Python and MATLAB.
Navigating the complex landscape of neural networks requires not only a firm grasp of theoretical foundations but also the practical skills to implement them effectively. Practical Neural Networks in Python and MATLAB is designed to be a definitive resource, offering a unique dual-language approach to mastering these powerful models.
Key Features:
A Dual-Language, Integrated Approach: This book provides a side-by-side exploration of neural networks in both Python and MATLAB. This methodology allows you to leverage Python's rich deep learning ecosystem (TensorFlow, Keras, PyTorch) and MATLAB's specialized toolboxes, giving you the flexibility to work within your preferred environment or across different project requirements.
Comprehensive Coverage of Algorithms and Architectures: Move beyond basic backpropagation. The text provides a systematic review of fundamental and advanced training algorithms, including Gradient Descent, Newton's Method, Levenberg-Marquardt, Recursive Least Squares (RLS), and metaheuristics like Genetic Algorithms and Particle Swarm Optimization. Furthermore, it offers a detailed survey of over 25 major neural network architectures, from foundational Perceptrons and Feedforward Networks to advanced systems like CNNs, RNNs (LSTM, GRU), Autoencoders, GANs, and Deep Belief Networks.
Practical, Code-Oriented Learning: Each concept and architecture is accompanied by ready-to-run code examples. This practical focus ensures that you can immediately translate theoretical understanding into functional code, experiment with parameters, and adapt the implementations to your own unique challenges.
Real-World Application and Case Studies: The learning is grounded in practicality through diverse case studies across multiple domains. You will find applications in medical diagnostics (e.g., diabetes classification), time-series forecasting (e.g., air quality prediction), system identification, natural language processing, and more. These examples provide complete pipelines from data preprocessing and model training to evaluation and visualization.
This Book is Ideal For:
University students and researchers in Computer Science, Artificial Intelligence, Engineering, and related fields.
R&D engineers and scientists working in algorithm development, data analysis, and intelligent systems.
Any practitioner seeking a thorough, hands-on understanding of neural networks with the flexibility to work in both Python and MATLAB environments.
In essence, Practical Neural Networks in Python and MATLAB serves as an invaluable companion for anyone looking to deepen their expertise in neural networks. It is more than a textbook; it is a practical toolkit designed to accelerate your research, enhance your projects, and provide a clear, comprehensive reference for the key architectures and algorithms shaping the field of AI today.
Table of Contents:
Introduction.- Multilayer Perceptron (MLP) Neural Networks.- Recursive Least Squares (RLS) Based Neural Network Training.- Neural Networks Training Based on Second-Order Optimization Technique.- Neural Network Training Based on Genetic Algorithm.- Neural Network Training Based on Particle Swarm Optimization (PSO).- UKF-based Neural Network Training.- Introduction of Machine learning libraries in Python with illustrative examples.- A summary of different type of neural networks in Matlab and Python.- Bibliography.
About the Author :
Prof. Dr. Chunwei Zhang is the Chair Distinguished Professor at Shenyang University of Technology, China. He is the Founding Director of the Multidisciplinary Center for Infrastructure Engineering at Shenyang University of Technology, and the Founding Director of the Structural Vibration Control Research Group at Qingdao University of Technology. His research achievement and worldwide impact have been highly recognized by the international academia society, as evidenced by the continuous inclusions into the top global rankings, e.g. Clarivate, Elsevier, and Stanford etc. His inventions have been implemented into engineering practice, e.g. the active control system for the Canton Tower etc. He has received many prestigious national and international awards for excellence in research. His research area includes Engineering, AI, Mechanics, Materials and many crossing/inter-disciplines.
Mr. Tianpeng Li is a PhD student at Shenyang University of Technology, China. His research area is active control for structures.
Ms. Ying Dai is a PhD student at Northeastern University, China. Her research area is laser illumination, inorganic luminescent materials and simulations.
Prof. Dr. Li Sun is appointed as the Distinguished Professor at Shenyang Jianzhu University by the Ministry of Education of China. She was the Visiting Professor at Curtin University (Australia), Nanyang Technological University (Singapore), and Hong Kong Polytechnic University etc. She has received many prestigious academic titles, including the China Bai-Qian-Wan Talent (Bai Level), Endeavour Fellow of Australia, Distinguished Professor of Liaoning Province, Leader of Innovation Teams of Liaoning Provincial Universities, Outstanding Teacher of Liaoning Province etc. She has won many national, provincial and ministerial level scientific and technological awards as leaders or participants.
Prof. Dr. Ardashir Mohammadzadeh is a Professor and Supervisor of graduate students at Sakarya University, and Shenyang University of Technology, in the field of intelligent control systems. He earned his PhD from University of Tabriz, Azerbaijan, Iran, in 2016. Over his career, Prof. Mohammadzadeh has made substantial contributions to advanced fuzzy logic systems, hybrid control strategies, fault-tolerant control for complex nonlinear and cyber-physical systems, and intelligent systems/controllers for medical applications.
