Jour Fixe - XAI-Usage in the Financial Sector

 XAI-Usage in the Financial Sector (joint with K. Valerie Carl, Domenic Kellner, Patrick Weber, and Prof. Dr. Oliver Hinz)

The continuous developments in the field of Artificial Intelligence (AI) and Machine Learning (ML) have led to various AI-based applications in business and private life. Many ML models propose classifications based on a “black-box” principle, i.e. these applications only reveal input and output but not the connection between them. Other ML models such as logistic regressions or decision trees can clearly explain how they behave, how they produce predictions and what the influencing variables are. However, this type of model usually comes with lower prediction accuracy. Newer, so-called eXplainable AI methods are developed to align these two types.

As AI can be deployed to support even high-impact decisions like in the financial sector, comprehensible decisions matter. Due to the high degree of required transparency and accountability in the financial sector, AI-deployment in this specific sector is not as simple and therefore also not as common as it is in some other industries. Still, only few high-quality research projects focus on AI models that meet the minimum requirements of regulatory authorities like the German BaFin or the European EBA  and at the same time customer requirements. Therefore, research providing actionable AI models for banking is scare.  In this talk, we will present an overview of our current research efforts regarding XAI-applicability in finance.