Tutorial 2: SoC Architectures for accelerating AI Computing

Tutorial 2: SoC Architectures for accelerating AI Computing
Speaker: Bharat Daga (Intel)
Tutorial Abstract:
The proliferation of AI in various fields has resulted in exponential increase in the compute. This necessitates to rethink the compute architecture to keep pace with the demand for compute, memory bandwidth and inter process communication. In this tutorial, we will explore the various compute architectural solutions targeting AI workloads. The tutorial will delve into custom accelerators and various tradeoffs for performance and power.
About Speaker:
Bharat Daga is a Senior Principal Engineer and Director of Engineering, Habana Labs, Intel. He has an industry experience of over 24 Years in the field of Computer Architecture, AI/ML Hardware, CPU and VLSI design. For the last 8 years he has incubated and lead a design of AI/ML compute core targeting data centers and servers. He is the recipient of IAA - Intel Achievement Award the highest award in Intel for his pathbreaking work on AI/ML products. Bharat has a Masters in Computer Engineering from University of Delaware, USA and a bachelors from Osmania University, Hyderabad. He is a great sports enthusiast and regularly plays volleyball to keep fit.