Sivanandam breaks down the complex mathematics of biological and artificial neurons into structured, digestible modules. Biological vs. Artificial Neurons
The simplest form of a feedforward network. The book demonstrates its limitation in solving non-linearly separable problems (like the XOR gate).
: Using commands like newff to define structure and initialize weights.
: Each chapter includes summaries and review questions tailored for semester-based exam preparation. Availability & Format
Engineers and students looking for the typically use it as a companion guide for legacy systems validation or academic reference. Sivanandam breaks down the complex mathematics of biological
There are several types of neural networks, including:
Do you need help to run on a modern version of MATLAB?
net = newff([min_val max_val], [hidden_neurons output_neurons], 'tansig' 'purelin', 'traingd'); Use code with caution.
: If you try to run Sivanandam's MATLAB 6.0 code on a modern version of MATLAB (e.g., R2026a), you will encounter errors. Functions like newff have been replaced by feedforwardnet , and sim is often bypassed by calling the network object directly as a function (e.g., net(P) ). The book demonstrates its limitation in solving non-linearly
: Single-layer and a brief introduction to multi-layer networks.
Sivanandam’s text dedicates significant focus to the Backpropagation Network (BPN). BPNs utilize gradient descent to minimize the Mean Squared Error (MSE) between predicted outputs and actual targets. In MATLAB 6.0, a BPN was initialized using newff :
The students groaned. Riya crossed her arms.
The text details several critical neural network models that are essential for beginners: Availability & Format Engineers and students looking for
Introduction to Neural Networks using MATLAB 6.0 S.N. Sivanandam
: It covers basic building blocks including neurons , weights , biases , and activation functions .
When utilizing digital copies or library resources for this textbook, keep these tips in mind: