78 citations · 341 across the 20 of their papers we have counts for
3 papers · 1 filter
A two-head loss function for deep Average-K classification
Camille Garcin, Maximilien Servajean, Alexis Joly +1
Average-K classification is an alternative to top-K classification in which the number of labels returned varies with the ambiguity of the input image but must average to K over al…
Identify ambiguous tasks combining crowdsourced labels by weighting Areas Under the Margin
Tanguy Lefort, Benjamin Charlier, Alexis Joly +1
In supervised learning - for instance in image classification - modern massive datasets are commonly labeled by a crowd of workers. The obtained labels in this crowdsourcing settin…
Optimal checkpointing for heterogeneous chains: how to train deep neural networks with limited memory
Julien Herrmann, Olivier Beaumont, Lionel Eyraud-Dubois +3
This paper introduces a new activation checkpointing method which allows to significantly decrease memory usage when training Deep Neural Networks with the back-propagation algorit…