Vortrag "Erkennung von Abweichungen jeglicher Art (Anomalieerkennung)" bei Veranstaltung der KI-Allianz Rheinland-Pfalz in Ludwigshafen

• Event Location: Ludwigshafen

  Event Date: 29.11.2022

  Contact Person: Prof. Dr. Marius Kloft

  

  Prof. Marius Kloft, "Detection of Deviations of Any Kind (Anomaly Detection)"

  Anomalies (deviations from the norm) are absolutely ubiquitous, from attacks on computer networks to critical errors in a chemical factory or rare tumors in histopathological images. How can such (and other anomalies) be detected? This question has plagued mankind for thousands of years. In his talk, Marius Kloft presents the current state of the art in anomaly detection and illustrates how modern anomaly detection methods can be profitably used to solve practical problems in science and engineering.

  Possible application scenarios:

  • Chemical industry (e.g., monitoring critical process conditions)
  • Irregularities in images/videos (e.g. detection of unexpected objects)
  • Manufacturing (e.g. monitoring of tolerance ranges)
  • Maintenance (e.g., predicting failure)
  • Medicine (e.g., progress monitoring, rare disease detection)
  • IT security (e.g. detection of attacks on computer systems and networks)
  • Financial accounting (e.g. control of bank transfers)
    
    

  Prof. Marius Kloft has researched and taught at various institutions in Germany and the USA (including TU Berlin, UC Berkeley, USC Los Angeles, New York University, Memorial Sloan-Kettering Cancer Center New York, and Google Research). Since 2017, he has been a professor of machine learning at TU Kaiserslautern. His research ranges from the development of new learning algorithms to applications in process engineering, medicine, and mechanical engineering.

  29.11.2022, 13:30 - 16:00, "Freischwimmer-Club" Ludwigshafen - an event of the AI Alliance Rhineland-Palatinate together with the IHK Pfalz

  

  Prof. Dr. Marius explains "Deep Anomaly Detection" in his presentation