IPU-M2000 and IPU-POD: New Breakthroughs in AI at scale

Introducing our next generation machine intelligence scale-out systems

Webinar - IPU Systems for AI at Scale

Machine learning is undergoing a rapid evolutionary phase. In areas such as computer vision and natural language processing, the pace of change has been particularly striking. Despite this, future progress is being restricted by existing architectures which are not built to adapt to new directions of model evolution.

This webinar will look at some of the challenges machine learning engineers face as this new world of evolutionary models becomes the norm. Graphcore's radical new approach to machine intelligence - compute, communications and data - will give innovators the flexibility to embrace this shift and make unprecedented breakthroughs.

What you'll learn:

  • Why compute, data efficiency and communications are integral to enabling innovation in machine intelligence
  • Insights into Graphcore's new generation of scale-out products: IPU-Machine: M2000 & IPU-POD64
  • How Graphcore's MK2 IPU architecture facilitates the research and deployment of new models
  • How IPU systems enhance model deployability and efficiency


Register to watch on-demand

Screen Shot 2020-07-13 at 19.45.12

Meet the speaker

Tom Wilson

Tom has almost 30 years of experience in the semiconductor and processor industry, the last 9 years in various advanced algorithm processing applications. Tom is VP Product Management at Graphcore responsible for product line management.  Tom has been involved in multiple startups as well as larger organizations and has a background in processor acceleration for such functions as computer vision and automotive radar processing.  As part of his role at Graphcore,  Tom presents at a range of conferences as part of communicating Graphcore’s revolutionary approach for machine intelligence processing.