Explore a 29-minute conference talk from the Toronto Machine Learning Series (TMLS) on Deep XVA, presented by Andrew Green. Delve into the world of XVA models, known for their computational intensity in finance, and discover how they leverage acceleration techniques like GPU computation and Adjoint Algorithmic Differentiation (AAD). Learn how deep learning offers a computationally efficient and implementation-friendly approach to approximating derivative valuation functions, a crucial element in XVA models. Gain insights into the seamless integration of deep learning with traditional quantitative finance models, resulting in an effective XVA calculation platform.