Abstract: Current deep learning methods are based on the repeated, expensive application of convolutions with parameter-intensive weight matrices. In this work, we present a novel concept that enables the application of differentiable random ferns in end-to-end networks. It can then be used as multiplication-free convolutional layer alternative in deep network architectures. Our experiments on the binary classification task of the TUPAC'16 challenge demonstrate improved results over the state-of-the-art binary XNOR net and only slightly worse performance than its 2x more parameter intensive floating point CNN counterpart.
Paper Type: methodological development
Track: short paper
Keywords: End-To-End Trainable Ferns, Network Efficiency, Binary Embedding
TL;DR: We demonstrate how to implement differentiable random ferns as energy & parameter efficient alternative to convolutional layers.
Presentation Upload: zip
Presentation Upload Agreement: I agree that my presentation material (videos and slides) will be made public.