IMPROVING THE TRUSTWORTHINESS OF AI EXPLANATIONS

Summary from the TLDR AI newsletter

This project improves the Model Parameter Randomisation Test, a key tool in eXplainable AI (XAI), by introducing two new versions: Smooth MPRT and Efficient MPRT.