Robust Unsupervised Learning via L-Statistic Minimization
Andreas Maurer 1 Daniela A. Parletta 1 2 Andrea Paudice 1 3 Massimilano Pontil 1 4
Abstract restrict attention to “a sufficient portion of the data in good
Designing learning algorithms that are resistant to agreement with one of the hypothesized models”.
perturbations of the underlying data distribution To implement the above idea, we propose using L-
is a problem of wide ...


雷达卡




京公网安备 11010802022788号







