Textual noise, such as typos or abbreviations, is a well-known issue tha...
Product matching is a fundamental step for the global understanding of
c...
Recently, the use of synthetic datasets based on game engines has been s...
Long iterative training processes for Deep Neural Networks (DNNs) are
co...
A recurring focus of the deep learning community is towards reducing the...
Deep neural networks trained with standard cross-entropy loss memorize n...
Self-supervised representation learning can mitigate the limitations in
...
Reducing the amount of labels required to trainconvolutional neural netw...
Noisy labels are an unavoidable consequence of automatic image labeling
...
Semi-supervised learning, i.e. jointly learning from labeled an unlabele...
This paper presents a novel approach for segmenting moving objects in
un...
Despite being robust to small amounts of label noise, convolutional neur...