Research Seminar on AI: Logical Scenarios for Systematic Training and Validation of AI-based Systems

 

Tuesday, March 14, 2023, 4:00pm

Simulations and synthetic data are a necessary supplement to real-world experiments in order to alleviate their effort, cost and risks. When dealing with massive parallel simulations exploring variants of operational scenarios, formalization of the operational scenario and its variation space opens various paths towards systematization of the quality- and safety-assurance of synthetic data and the artificial intelligence based system under test (SUT). This talk will discuss our approach towards formalization of simulation scenarios and their variation spaces as well as their potential in seamless integration of novel synthetic data-generation methodologies with systematic quality assurance.

Osama Maqbool received his M.Sc. degree in electrical engineering with a focus on automation and robotics in the Technical University of Dortmund in 2017. Since 2020, he is a PhD candidate at the Institute of Man-Machine-Interaction at RWTH Aachen investigating massive parallel simulation methods in order to generate diverse training datasets and validation tests for Artificial Intelligence based systems.