Experiments are implementing and are based on Chen Lee, P.W.Gallagher, Z.Tu, Gneralizing Pooling Functions in Convolutional Neural Networks: Mixed, Gated and Tree, 2015. My work was used to satisfy the coursework final project for the research course Machine Learning Practial in University of Edinburgh.
Experiment results again show the advantages obtained from these innovate pooling functions (mixed and gated) in CNN-like architectures in object recognition.
