Interpretable Self-Explanatory Machine Learning Models

Explore the critical importance of interpretable and self-explanatory machine learning models in this 32-minute conference talk from the Toronto Machine Learning Series. Delve into the challenges of model risk management, particularly for complex ML models in mission-critical or regulated applications. Examine the limitations of current machine learning explainability techniques and understand why inherently interpretable models are crucial. Learn about strategies to transform sophisticated neural networks and deep learning models into self-explanatory systems. Gain insights from Agus Sudjianto, EVP and Head of Corporate Model Risk at Wells Fargo, as he discusses the potential financial and non-financial harm caused by incorrect models and emphasizes the need for better understanding, testing, and managing of model failures and their unintended consequences.