This is an implementation of the Min-Max Modular Network based on Prof. Lu Bao Liang’s work Task Decomposition Based on Class Relations: a Modular Neural Network Architecture for Pattern Classification. Link to Paper
The implemented multi-class classifier used libsvm as the underlying classifier for smaller modules. The main idea is to use min-max strategies to combine result of multiple classifiers, circumventing the problem that skewed datasets are often met in classification tasks.
- libsvm 3.22