[HTML][HTML] Dual adversarial attacks: Fooling humans and classifiers

J Schneider, G Apruzzese - Journal of Information Security and Applications, 2023‏ - Elsevier
Adversarial samples mostly aim at fooling machine learning (ML) models. They often involve
minor pixel-based perturbations that are imperceptible to human observers. In this work …

Computing Efficient Data Summaries

S Mair - 2022‏ - pubdata.leuphana.de
Extracting meaningful representations of data is a fundamental problem in machine
learning. Those representations can be viewed from two different perspectives. First, there is …

Berechnung effizienter Datenzusammenfassungen

S Mair - D22, 2022‏ - dl.gi.de
Zusammenfassung Das Extrahieren sinnvoller Repräsentationen von Daten ist ein
grundlegendes Problem im maschinellen Lernen und kann aus zwei unterschiedlichen …