Deep learning is emerging as a major application for high-performance computing. While training of deep neural networks (DNNs) places some unique demands on compu"> Deep learning is emerging as a major application for high-performance computing. While training of deep neural networks (DNNs) places some unique demands on compu" /> Deep learning is emerging as a major application for high-performance computing. While training of deep neural networks (DNNs) places some unique demands on compu"> The Convergence of HPC and Deep Learning | Supercomputing 2016
 

GTC ON-DEMAND

 
SEARCH SESSIONS
SEARCH SESSIONS

Search All
 
Refine Results:
 
Year(s)

SOCIAL MEDIA

EMAIL SUBSCRIPTION

 
 

GTC ON-DEMAND

HPC and Supercomputing
Presentation
Media
The Convergence of HPC and Deep Learning
Abstract:

Deep learning is emerging as a major application for high-performance computing. While training of deep neural networks (DNNs) places some unique demands on computing hardware its shares with mainstream HPC applications the need for high performance arithmetic, high memory bandwidth, and high-bandwidth, low-latency networks. Deep learning can also be used to enhance traditional HPC applications both by interpreting the results and by “learning” constituent equations. This talk will examine the common requrements of DL and HPC and applications of DL to HPC. 

 
Topics:
HPC and Supercomputing, Artificial Intelligence and Deep Learning
Type:
Talk
Event:
Supercomputing
Year:
2016
Session ID:
SC6119
Streaming:
Share: