Skip to main content
  • A tutorial on automatic differentiation for scientific design: practical, elegant and powerful

Talk abstract:

Automatic differentiation (AD) is a numerical technique for computing the derivative of a function specified as a computer program. Although AD was invented decades ago, it wasn’t until the recent interest in machine learning and the associated development of high-quality automatic differentiation frameworks that the benefits of AD in physics were more widely recognized. In this tutorial, I introduce AD. By the end of the tutorial, you will hopefully understand the fundamentals of how AD works in theory and how it is used in practice. For a short, 5-minute introduction to AD, feel free to read this https://twitter.com/NMcgreivy/status/1351706692317138945?s=20 and this https://twitter.com/NMcgreivy/status/1286057985987563525?s=20.

About FusionEPtalks

FusionEPtalks is brought to you by students and researchers involved in the field of nuclear fusion. Our mission is to do student-led webinars, expert talks and panels on the development of nuclear fusion as an energy source that connect scholars, engineers and enthusiasts around the world.

 

 

A tutorial on automatic differentiation for scientific design: practical, elegant and powerful

Author/Publisher
Author/Publisher
FusionEPtalks
Other link