DLS: A Fast and Flexible Neural Network Training System With Fine-grained Heterogeneous Device Orchestration-Reference-Cited by-同舟云学术

DLS: A Fast and Flexible Neural Network Training System With Fine-grained Heterogeneous Device Orchestration

Author:

Park Pyeongsu¹^ORCID,Lee Jaewon²,Jeong Heetaek¹,Kim Jangwoo¹^ORCID

Affiliation:

1. Department of Electrical and Computer Engineering, Seoul National University, Seoul, South Korea

2. Facebook, Menlo Park, CA, USA

Funder

Samsung

Seoul National University

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Subject

Computational Theory and Mathematics,Hardware and Architecture,Signal Processing

Link

Reference37 articles.

1. Device placement optimization with reinforcement learning;mirhoseini;Proc Int Conf Mach Learn,2017

2. A hierarchical model for device placement;mirhoseini;Proc Int Conf Learn Representations,2018

3. Tensorflow: A system for large-scale machine learning.;abadi;Proc Symp Oper Syst Des Implementation,2016

4. Intune automatically compensated digital PoL controller with driver and PMBus telemetry.,0

Cited by 1 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献