Robust deep learning from weakly dependent data

arXiv:2405.05081v1 Announce Type: cross
Abstract: Recent developments on deep learning established some theoretical properties of deep neural networks estimators. However, most of the existing works on this topic are restricted to bounded loss functions or (sub)-Gaussian or bounded input. This paper considers robust deep learning from weakly dependent observations, with unbounded loss function and unbounded input/output. It is only assumed that the output variable has a finite $r$ order moment, with $r >1$. Non asymptotic bounds for the expected excess …

abstract arxiv cs.lg data deep learning function functions however loss math.st networks neural networks paper robust stat.ml stat.th type