Timezone: »
Despite the success of CNNs, selecting the optimal architecture for a given task remains an open problem. Instead of aiming to select a single optimal architecture, we propose a ``fabric'' that embeds an exponentially large number of architectures. The fabric consists of a 3D trellis that connects response maps at different layers, scales, and channels with a sparse homogeneous local connectivity pattern. The only hyper-parameters of a fabric are the number of channels and layers. While individual architectures can be recovered as paths, the fabric can in addition ensemble all embedded architectures together, sharing their weights where their paths overlap. Parameters can be learned using standard methods based on back-propagation, at a cost that scales linearly in the fabric size. We present benchmark results competitive with the state of the art for image classification on MNIST and CIFAR10, and for semantic segmentation on the Part Labels dataset.
Author Information
Shreyas Saxena (INRIA)
Jakob Verbeek (INRIA)
More from the Same Authors
-
2021 Spotlight: Instance-Conditioned GAN »
Arantxa Casanova · Marlene Careil · Jakob Verbeek · Michal Drozdzal · Adriana Romero Soriano -
2021 Poster: Instance-Conditioned GAN »
Arantxa Casanova · Marlene Careil · Jakob Verbeek · Michal Drozdzal · Adriana Romero Soriano -
2021 Poster: XCiT: Cross-Covariance Image Transformers »
Alaaeldin Ali · Hugo Touvron · Mathilde Caron · Piotr Bojanowski · Matthijs Douze · Armand Joulin · Ivan Laptev · Natalia Neverova · Gabriel Synnaeve · Jakob Verbeek · Herve Jegou -
2019 Spotlight: Adaptive Density Estimation for Generative Models »
Thomas Lucas · Konstantin Shmelkov · Karteek Alahari · Cordelia Schmid · Jakob Verbeek -
2010 Workshop: Beyond classification: Machine Learning for next generation Computer Vision challenges »
Craig Saunders · Jakob Verbeek · Svetlana Lazebnik -
2007 Oral: Scene Segmentation with CRFs Learned from Partially Labeled Images »
Jakob Verbeek · Bill Triggs -
2007 Poster: Scene Segmentation with CRFs Learned from Partially Labeled Images »
Jakob Verbeek · Bill Triggs