Abstract: This paper reports experimental results as the proof of concept to a novel approach for noncontact manipulation of magnetic objects using a combination of permanent magnets and mechanical ...
Abstract: Deep learning models are highly susceptible to adversarial attacks, where subtle perturbations in the input images lead to misclassifications. Adversarial examples typically distort specific ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results