no code implementations • 5 Mar 2019 • Ismaïla Seck, Gaëlle Loosli, Stephane Canu
Adversarial examples are a challenging open problem for deep neural networks.
no code implementations • 27 Feb 2019 • Gaëlle Loosli
The proposed work aims at proposing a alternative kernel decomposition in the context of kernel machines with indefinite kernels.
no code implementations • 11 Jun 2018 • Ismaïla Seck, Khouloud Dahmane, Pierre Duthon, Gaëlle Loosli
This paper presents the recently published Cerema AWP (Adverse Weather Pedestrian) dataset for various machine learning tasks and its exports in machine learning friendly format.
no code implementations • 7 Dec 2017 • Gaëlle Loosli, Hattoibe Aboubacar
This paper proposes the adaptation of Support Vector Data Description (SVDD) to the multiple kernel case (MK-SVDD), based on SimpleMKL.