Open Peer Review. Open Publishing. Open Access. Open Discussion. Open Directory. Open Recommendations. Open API. Open Source.
Fast Training of Convolutional Networks through FFTs
Michael Mathieu, Mikael Henaff, Yann LeCun
Dec 24, 2013 (modified: Dec 24, 2013)ICLR 2014 conference submissionreaders: everyone
Decision:submitted, no decision
Abstract:Convolutional networks are one of the most widely employed architectures in computer vision and machine learning. In order to leverage their ability to learn complex functions, large amounts of data are required for training. Training a large convolutional network to produce state-of-the-art results can take weeks, even when using modern GPUs. Producing labels using a trained network can also be costly when dealing with web-scale datasets. In this work, we present a simple algorithm which accelerates training and inference by a significant factor, and can yield improvements of over an order of magnitude compared to existing state-of-the-art implementations. This is done by computing convolutions as pointwise products in the Fourier domain while reusing the same transformed feature map many times. The algorithm is implemented on a GPU architecture and addresses a number of related challenges.
Enter your feedback below and we'll get back to you as soon as possible.