Skip to yearly menu bar Skip to main content


DISCO Nets : DISsimilarity COefficients Networks

Diane Bouchacourt · Pawan K Mudigonda · Sebastian Nowozin

Area 5+6+7+8 #11

Keywords: [ Deep Learning or Neural Networks ]


We present a new type of probabilistic model which we call DISsimilarity COefficient Networks (DISCO Nets). DISCO Nets allow us to efficiently sample from a posterior distribution parametrised by a neural network. During training, DISCO Nets are learned by minimising the dissimilarity coefficient between the true distribution and the estimated distribution. This allows us to tailor the training to the loss related to the task at hand. We empirically show that (i) by modeling uncertainty on the output value, DISCO Nets outperform equivalent non-probabilistic predictive networks and (ii) DISCO Nets accurately model the uncertainty of the output, outperforming existing probabilistic models based on deep neural networks.

Live content is unavailable. Log in and register to view live content