Imagine if an AI model could not only make predictions but also provide a clear measure of its own uncertainty. As AI becomes increasingly integrated into critical applications, understanding the confidence and reliability of predictions is essential for building trust in these systems. Conformal Prediction (CP) offers a robust, statistically grounded approach to uncertainty quantification that can be applied to virtually any model, regardless of the underlying data distribution. In this talk, we will explore the core principles of CP and demonstrate how it can be applied to practical scenarios, such as time series forecasting and visual inspection tasks.
Principal Data Scientist, Boston Scientific Corporation
Sruthi Pisipati, Principal Data Scientist at Boston Scientific, leverages AI to drive innovation in medical device technology. Her interest areas include Vision, Gen AI and Forecasting domains.
Principal Data Scientist, Boston Scientific Corporation
Jason’s path has wound through chemistry, materials science, computer science, and electron microscopy. Now a data scientist, he specializes in AI vision and data-driven design of materials.
