Backpropagation Algorithm for Three-Layer Feedforward Neural Network

Implementation of the Backpropagation (BP) algorithm for a three-layer feedforward neural network. The program includes the following key features: (1) Configurable node counts for each layer (input, hidden, output); (2) Adjustable learning rate η for controlling weight update speed; (3) Weight initialization with random values in the [-1, 1] range; (4) Support for both unipolar and bipolar Sigmoid activation functions.

Generation Model of Scale-Free Networks

This universal source code implementation generates scale-free networks with configurable node counts, featuring preferential attachment algorithms and degree distribution visualization capabilities.