HPC and AI - Two Sides of the Same Coin

AI has existed for decades, but it is only HPC power that makes AI useful in real-life situations. This success led to further improvements in AI algorithms for solving complex problems, such as designing chips and making systems faster. In other words, AI is possible because of HPC.

AI, in turn, has enabled faster systems and efficient hardware architectures. By automating EDA processes involved in chip design, AI motivates the HPC community to explore novel technologies beyond Moore's law to build efficient deep learning models. With HPC feeding AI, and AI feeding HPC, we get better computing systems and better AI models – but will this ever stop?

Agenda:
A talk on “Accelerating AI for Science with a Reconfigurable Dataflow Architecture" by Dr. Raghu Prabhakar, Senior Principal Engineer, SambaNova Systems.

Followed by a fireside chat on “HPC and AI: How far can they take us together?” with
Dr. Preeti Malakar, Assistant Professor, IIT Kanpur;
Dr. Raghu Prabhakar, Senior Principal Engineer, SambaNova Systems;
Dr. Yogesh Simmhan, Associate Professor, IISc Bangalore;
Dr. Sparsh Mittal, Assistant Professor, IIT Roorkee along with the moderator
Dr. Rekha Singhal, Principal Scientist, TCS Research.