GTC ON-DEMAND

 
SEARCH SESSIONS
SEARCH SESSIONS

Search All
 
Refine Results:
 
Year(s)

SOCIAL MEDIA

EMAIL SUBSCRIPTION

 
 

GTC ON-DEMAND

Presentation
Media
Abstract:
AI has fueled a paradigm shift in web search during the past few years, transforming the user experience from simple keyword-based queries in a browser to the rich multimodal voice, vision, and context-based searches of today. We'll discuss approaches we used in Microsoft Bing to drive this shift at hyperscale and provide a detailed overview of sophisticated deep learning models and software techniques used in production. We'll also describe the scalable underlying platform that unlocks the power of the latest hardware, including GPUs. Our talk will highlight some of the lessons learned from driving this transformation while simultaneously operating a production service at large scale.
AI has fueled a paradigm shift in web search during the past few years, transforming the user experience from simple keyword-based queries in a browser to the rich multimodal voice, vision, and context-based searches of today. We'll discuss approaches we used in Microsoft Bing to drive this shift at hyperscale and provide a detailed overview of sophisticated deep learning models and software techniques used in production. We'll also describe the scalable underlying platform that unlocks the power of the latest hardware, including GPUs. Our talk will highlight some of the lessons learned from driving this transformation while simultaneously operating a production service at large scale.  Back
 
Topics:
Deep Learning & AI Frameworks
Type:
Talk
Event:
GTC Silicon Valley
Year:
2019
Session ID:
S9138
Streaming:
Download:
Share:
 
Abstract:
A common challenge for large-scale online services such as Microsoft Bing or Cortana is ensuring that queries complete fast.  Additionally, these services require support for new models to be created, tested and deployed in production at a very fast pace.  Come and learn some of the techniques we have used to deliver both.  Bing Search and Cortana adopt advanced Machine Reading Comprehension (MRC) models for answer extraction and identifying answer relevance. We will describe a number of innovations in the MRC models and also techniques for hardware acceleration at production scale.  
A common challenge for large-scale online services such as Microsoft Bing or Cortana is ensuring that queries complete fast.  Additionally, these services require support for new models to be created, tested and deployed in production at a very fast pace.  Come and learn some of the techniques we have used to deliver both.  Bing Search and Cortana adopt advanced Machine Reading Comprehension (MRC) models for answer extraction and identifying answer relevance. We will describe a number of innovations in the MRC models and also techniques for hardware acceleration at production scale.    Back
 
Topics:
Speech & Language Processing
Type:
Talk
Event:
GTC Silicon Valley
Year:
2018
Session ID:
S81039
Streaming:
Share:
 
 
Previous
  • Amazon Web Services
  • IBM
  • Cisco
  • Dell EMC
  • Hewlett Packard Enterprise
  • Inspur
  • Lenovo
  • SenseTime
  • Supermicro Computers
  • Synnex
  • Autodesk
  • HP
  • Linear Technology
  • MSI Computer Corp.
  • OPTIS
  • PNY
  • SK Hynix
  • vmware
  • Abaco Systems
  • Acceleware Ltd.
  • ASUSTeK COMPUTER INC
  • Cray Inc.
  • Exxact Corporation
  • Flanders - Belgium
  • Google Cloud
  • HTC VIVE
  • Liqid
  • MapD
  • Penguin Computing
  • SAP
  • Sugon
  • Twitter
Next