Skip to content Skip to navigation

Sylvain Dubois

Emerging Memories for AI applications

Sylvain Dubois, Vice President of Strategic Marketing & Business Development, Crossbar Systems

Abstract:
Today, even the most advanced architectures are suffering limitations caused by the memory bottleneck between computing cores and data and the restrictions of traditional AI methodologies that depend on classification of data. The use of emerging memories with AI applications is a natural solution for these memory-intensive applications that will also need to be energy efficient.

While memory is the critical foundation of next-generation chip processors, industry collaboration is needed to enable new AI capabilities including massive multi-modal search acceleration and machine learning at the Edge. To accelerate the deployment of ready-made, power-efficient solutions with unsupervised learning and event recognition capability, Crossbar has partnered with Gyrfalcon Technology Inc., mtes Neural Networks Corporation (mNN) and Robosensing Inc to create the SCalable AI for Learning at the Edge (SCAiLE) consortium. These companies bring their expertise in neural network hardware accelerators, AI datasets optimization and system integration to advance the activities of the consortium.

The consortium addresses the restrictions of traditional AI methodologies that depend on classification of data. The huge growth of IoT systems, including thousands of remote edge devices such as sensor-equipped cameras, creates a torrent of unstructured information in multiple forms that pours into cloud-located servers that cannot be handled effectively by classification alone.

This presentation will discuss:

  • The AI memory bottleneck in today’s systems 
  • The collaboration needed among the industry and the progress of a new AI consortium (SCAiLE) that is working to deliver scalable and power-saving accelerators that use optimized neural networks at the edge to quickly analyze and respond to multi-modal information (including video, images, speech, keywords and sensor feeds). 
  • Use cases including work with Japanese authorities on infrastructure for the 2020 Olympics, including video-based event detection and response capability. 

Biography:
Sylvain Dubois joined the Crossbar management team in 2013 as Vice President of Strategic Marketing and Business Development. With over 17 years of semiconductor experience in business development and strategic product marketing, he brings a proven ability to analyze market trends, identify new, profitable business opportunities and create precise product positioning that is in perfect sync with market demands to drive market share leadership and business results.

Prior to joining Crossbar, Mr. Dubois led strategic product positioning and market engagement for developing new products at Spansion. Responsible for identifying new growth opportunities and expanding the product portfolio, Mr. Dubois was instrumental in defining the Spansion Flash memory product roadmap.

From 2002-2006, Mr. Dubois was a System-on-Chip architect of OMAP application processors at Texas Instruments. Mr. Dubois was in charge of the architecture and technology roadmap of DRAM and Flash memory controllers, and developed strategic relationships with major DRAM and Flash memory suppliers.

Sylvain Dubois holds a Master of Science in Microelectronics from E.S.I.E.E. (Paris), University of Southampton (UK) and Universidad Pontifica Comillas (Spain).