2 papers
cs.MS2018
Dynamic Automatic Differentiation of GPU Broadcast Kernels
Jarrett Revels, Tim Besard, Valentin Churavy +2
We show how forward-mode automatic differentiation (AD) can be employed within larger reverse-mode computations to dynamically differentiate broadcast operations in a GPU-friendly…
cs.DC2016
High-level GPU programming in Julia
Tim Besard, Pieter Verstraete, Bjorn De Sutter
GPUs are popular devices for accelerating scientific calculations. However, as GPU code is usually written in low-level languages, it breaks the abstractions of high-level language…