604 citations · 2.5k across the 63 of their papers we have counts for
5 papers · 1 filter
Spatially Adaptive Computation Time for Residual Networks
Michael Figurnov, Maxwell D. Collins, Yukun Zhu +4
This paper proposes a deep learning architecture based on Residual Network that dynamically adjusts the number of executed layers for the regions of the image. This architecture is…
Fast Adaptation in Generative Models with Generative Matching Networks
Sergey Bartunov, Dmitry P. Vetrov
Despite recent advances, the remaining bottlenecks in deep generative models are necessity of extensive training and difficulties with generalization from small number of training…
Robust Variational Inference
Michael Figurnov, Kirill Struminsky, Dmitry Vetrov
Variational inference is a powerful tool for approximate inference. However, it mainly focuses on the evidence lower bound as variational objective and the development of other mea…
Ultimate tensorization: compressing convolutional and FC layers alike
Timur Garipov, Dmitry Podoprikhin, Alexander Novikov +1
Convolutional neural networks excel in image recognition tasks, but this comes at the cost of high computational and memory complexity. To tackle this problem, [1] developed a tens…
GTApprox: surrogate modeling for industrial design
Mikhail Belyaev, Evgeny Burnaev, Ermek Kapushev +4
We describe GTApprox - a new tool for medium-scale surrogate modeling in industrial design. Compared to existing software, GTApprox brings several innovations: a few novel approxim…