Explainable AI for Financial Services
Developing transparent machine learning models and governance frameworks to support responsible use of AI in lending and investment decisions.
View project
This project explores model interpretability techniques for credit scoring and transaction monitoring, working with partners to design explainability reports that can be understood by risk teams and regulators.
- Comparison of model-agnostic and model-specific methods.
- Human-in-the-loop evaluation with financial practitioners.
- Guidance for responsible deployment in production systems.