MAMMOth Project

MAMMOth focuses on multidiscrimination mitigation for tabular, network and multi-modal data. The project intends to address bias in AI through the following objectives: Redefine bias on the basis of multiple (protected) characteristics instead of a single attribute. Create standardised AI solutions to address bias across all phases of the development of AI systems. Develop and advance new technologies to evaluate and mitigate AI bias. Ensure reliability, traceability and explainability of AI solutions in order to address different types of bias. Integrate and prototype bias-preventing AI solutions into an open source suite in order to increase their availability and deployment. Increase awareness and skills preventing AI bias and the uptake of the MAMMOth solutions by involving affected communities and stakeholders across Europe and worldwide. Advance the European approach to excellence in trustworthy AI. Study AI biases on use cases – providing insights on high-risk applications of bias, in finance, face recognition and research.