Intelligent Systems Research Group
Q-AMeLiA
Quality Assurance of Machine Learning Applications
Team:
Alexander Melde, M.Sc.
Prof. Dr. Astrid Laubenheimer
Funding: MWK
Duration: 2020 - 2023
Overview
In Q-AMeLiA three Universities of Applied Sciences are developing suitable instruments for evaluating data quality. The focus here is on the representative coverage of the feature space and the evaluation of the quality of the AI model learned in the learning process. This safeguards the product risk of the manufacturer of AI-based products and guarantees the customer a quantified performance of the products with regard to the decisions of the AI.
The aim of the association thereby is to support the consortium consisting of five SMEs on the implementation of the special machine learning software development lifecycle (ML-SDLC) and the incorporation of important quality indicators.
Related Publications
Melde A., Gavrikov P., Madan M., Hoof D., Laubenheimer A., Keuper J., Reich C.
Tackling Key Challenges of AI Development – Insights from an Industry-Academia Collaboration
Upper-Rhine Artificial Intelligence Symposium UR-AI 2022, October 19th, 2022. [pdf] [poster]
Melde A., Laubenheimer A., Link N., Schauer C.
An Architecture to Quantify the Risk of AI-Models
Upper-Rhine Artificial Intelligence Symposium UR-AI 2021, October 27th, 2021 [pdf]
Partners & Funding
The research results are transferred to the private sector in close cooperation with the project partners:
The project is funded by the Ministry of Science, Research and Arts of Baden-Württemberg (MWK).
