February 23, 2020

147 words 1 min read

Modular convolution considered beneficial

Modular convolution considered beneficial

Jack Chung, Chao Liu, and Daniel Lowell explore breaking convolution algorithms into modular pieces to be better fused with graph compilers such as accelerated linear algebra (XLA).

Talk Title Modular convolution considered beneficial
Speakers Jack Chung (AMD), Chao Liu (AMD), Daniel Lowell (AMD)
Conference O’Reilly TensorFlow World
Conf Tag
Location Santa Clara, California
Date October 28-31, 2019
URL Talk Page
Slides Talk Slides

miOpen contains performance-critical GPU kernels that drive machine learning workloads on the AMD ROCm platform. Jack Chung, Chao Liu, and Daniel Lowell explore how to make them into modular pieces so they can be easily tuned for various GPU hardware from AMD and closely knitted with graph compilers such as TensorFlow XLA. They show how various convolution algorithms are implemented on AMD hardware, how they’re decomposed into modular pieces, how they can be picked up and fused by XLA, and how they perform.

comments powered by Disqus