Naveed Akram

Since the completion of his Master’s degree in Commercial Vehicle Technology at the University of Kaiserslautern-Landau (RPTU), Naveed Akram has been working at Fraunhofer IESE in the Safety Engineering department. He is continuing his work in the area of dependable machine learning, which he started during his studies. In this regard, he has worked in several national and international projects in which machine learning (ML) played a prominent role, including the projects SECREDAS, BIECO, SESAME, and SPELL, to name but a few. He has worked in a variety of safety-critical and finance-critical application areas, ranging from pedestrian/person detection (SESAME) to investment portfolio prediction (BIECO). He is currently overseeing the application and development of a tool called SafeML, which takes into account the out-of-context uncertainties resulting from ML models. -- Seit dem Abschluss seines Masterstudiums in Nutzfahrzeugtechnik an der RPTU Kaiserslautern-Landau arbeitet Naveed Akram am Fraunhofer IESE in der Abteilung Safety Engineering. Dort setzt er seine im Studium begonnene Arbeit im Bereich Verlässliches Maschinelles Lernen fort. In diesem Zusammenhang hat er an mehreren nationalen und internationalen Projekten mitgearbeitet, in denen Maschinelles Lernen (ML) eine wichtige Rolle spielte, darunter u.a. die Projekte SECREDAS, BIECO, SESAME und SPELL. Er hat in einer Vielzahl von sicherheits- und finanz-kritischen Anwendungsbereichen gearbeitet, von der Erkennung von Fußgängern/Personen (SESAME) bis zur Vorhersage von Anlageportfolios (BIECO). Derzeit leitet er die Anwendung und Entwicklung eines Tools namens SafeML, das die aus ML-Modellen resultierenden kontextunabhängigen Unsicherheiten berücksichtigt.

Dealing with uncertainties of Machine Learning components (Part 2)

Using Machine Learning components in critical systems requires a sound safety concept and the ability to argue and prove that the risk of the considered system is acceptably low. In our previous blog post (Dealing with uncertainties of Machine Learning…

Dealing with uncertainties of Machine Learning components (Part 1)

The use of Machine Learning (ML) components in safety-critical or financially critical systems is challenging. At Fraunhofer IESE, we address this challenge by systematically engineering comprehensive multi-layered safety concepts and explicitly considering sources of uncertainties. This specifically includes situations at…