Abstract - Strong demands for efficient deployment of Deep Learning (DL) applications prompt the rapid development of a rich DL ecosystem. To keep up with its fast advancement, it is crucial for DL frameworks to efficiently integrate a variety of optimized libraries and runtimes as their backends and generate the fastest possible executable by using them properly. However, current DL frameworks require significant manual effort to integrate diverse backends and often fail to deliver high performance. In this work, we propose Collage, an automatic framework for integrating DL backends. Collage provides a backend registration interface that allows users to precisely specify the capability of various backends. By leveraging the specifications of available backends, Collage searches for an optimized backend placement for a given workload and execution environment. Our evaluation shows that Collage automatically integrates multiple backends together without manual intervention, and outperforms existing frameworks by 1.21x, 1.39x, 1.40x on two different NVIDIA GPUs and an Intel CPU respectively.

Zoom 회의 링크 : https://snu-ac-kr.zoom.us/j/84866231819?pwd=TFoyY3A5VEtMazhSVnh4WHYzKzB1Zz09 (암호: snuspl)

Bio - Byungsoo Jeon is a Ph.D. student in the Computer Science Department at Carnegie Mellon University. He is advised by Prof. Tianqi Chen and Prof. Zhihao Jia. He is also a member of the Catalyst research group. His research interest lies in the intersection of Machine Learning and Computer Systems. He believes there are plenty of exciting challenges in automating sequential decision-making in learning systems. He is passionate about tackling these challenges by co-designing algorithms and learning systems. To this end, he works towards building a learning system that automates cross-stack optimizations.

문의: 소프트웨어플랫폼연구실(02-880-1611)