Tim Gymnich
Tim is a postgraduate student at TU Munich, where he is studying for his M.Sc. in Computer Science. His main interests are in compiler optimizations for high performance computing and programming languages.
Session
07-27
13:00
30min
Fast Forward and Reverse-Mode Differentiation via Enzyme.jl
Valentin Churavy, William Moses, Ludger Paehler, Tim Gymnich
Enzyme is a new LLVM-based differentiation framework capable of creating fast derivatives in a variety of languages. In this talk we will showcase improvements in Enzyme.jl, the Julia-language bindings for Enzyme that enable us to differentiate through parallelism (Julia tasks, MPI.jl, etc), mutable memory, JIT-constructs, all while maintaining performance. Moreover we will also showcase Enzyme's new forward mode capabilities in addition to its existing reverse-mode features.
JuliaCon
Purple