Recursive Autoconvolution for Unsupervised Learning of Convolutional Neural Networks

06/02/2016 ∙ by Boris Knyazev, et al. ∙ 0

In visual recognition tasks, such as image classification, unsupervised learning exploits cheap unlabeled data and can help to solve these tasks more efficiently. We show that the recursive autoconvolution operator, adopted from physics, boosts existing unsupervised methods by learning more discriminative filters. We take well established convolutional neural networks and train their filters layer-wise. In addition, based on previous works we design a network which extracts more than 600k features per sample, but with the total number of trainable parameters greatly reduced by introducing shared filters in higher layers. We evaluate our networks on the MNIST, CIFAR-10, CIFAR-100 and STL-10 image classification benchmarks and report several state of the art results among other unsupervised methods.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 1

page 2

Code Repositories

autocnn_unsup_py

Python scripts for paper "Recursive Autoconvolution for Unsupervised Learning of Convolutional Neural Networks" (https://arxiv.org/abs/1606.00611)


view repo
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.