Multilayer Perceptron Algebra
This work addresses the challenge of systematic neural network design for machine learning practitioners, though it appears incremental as it builds on existing MLP frameworks.
The authors tackled the problem of neural network design being reliant on developer intuition by introducing MLP algebra, a mathematical structure that provides a guiding principle for constructing MLPs tailored to specific datasets and building complex networks from simpler ones.
Artificial Neural Networks(ANN) has been phenomenally successful on various pattern recognition tasks. However, the design of neural networks rely heavily on the experience and intuitions of individual developers. In this article, the author introduces a mathematical structure called MLP algebra on the set of all Multilayer Perceptron Neural Networks(MLP), which can serve as a guiding principle to build MLPs accommodating to the particular data sets, and to build complex MLPs from simpler ones.