Adjoint Algorithmic Differentiation (AAD) Masterclass - New York
This course will provide a comprehensive understanding of the mathematical foundations of AAD and its role within financial applications.
This two-day course, provides a practical introduction to algorithmic differentiation (AD). Attendees will discuss the mathematical foundations for adjoints methods, algorithmic differentiation (AD) as a general computational technique for the efficient calculation of price sensitivities, and the use of AD software as a way to generate the adjoint code. Focus will be placed on its application to Monte Carlo methods for SDEs and finite difference methods for PDEs.
What Will You Learn?
- Comprehensive understanding of the mathematical foundations of AAD and its role within financial applications
- Hands-on case studies of application of AAD to Monte Carlo method for SDEs
- Hands-on case studies of application of AAD to finite difference method for PDEs
- Detailed understanding of second and higher-order AAD and of dco/c++ tool support
- Understanding of advanced issues in AAD including:
- methods for reducing the memory requirement of adjoint code
- symbolic adjoints of implicit functions including linear and nonlinear solvers and optimization methods
- handling and exploitation of shared and distributed memory parallelism in AAD
Who Should Attend?
Relevant departments may include but are not limited to:
- Quantitative Analysis
- Risk Management
- Fixed Income
- Equity / Credit / Commodity / FX Derivatives
- Portfolio Management
After the course participants will have gained knowledge of:
- Adjoints in Finance I
- Writing First-Order Adjoint Code by Hand
- Adjoints in Finance II
- Introduction to AAD Software Tool dco/c++
- Hands-on First-Order AAD with dco/++