Dynamic control flow in large-scale machine learning

EUROSYS '18: PROCEEDINGS OF THE THIRTEENTH EUROSYS CONFERENCE, pp. 18:1-18:15, 2018.

Cited by: 29|Bibtex|Views296|DOI:https://doi.org/10.1145/3190508.3190551
EI
Other Links: dblp.uni-trier.de|academic.microsoft.com|arxiv.org

Abstract:

Many recent machine learning models rely on fine-grained dynamic control flow for training and inference. In particular, models based on recurrent neural networks and on reinforcement learning depend on recurrence relations, data-dependent conditional execution, and other features that call for dynamic control flow. These applications ben...More

Code:

Data:

Your rating :
0

 

Tags
Comments