Pratik Kumar’s research paves a new way for designing and synthesizing scalable & reconfigurable analogue accelerators that are orders of magnitude more energy and area-efficient than conventional digital accelerators.
Falling Walls
Falling Walls Science Summit 2023
Login or Register
You need to be logged in to use this feature.
Reducing the Carbon Footprint of Artificial Intelligence: Breaking the Wall of Analog AI Technology
Pratik Kumar is a Doctoral Research Candidate at the Indian Institute of Science (IISc), NeuRonICS lab, Bengaluru, India. His research lies at the intersection of artificial intelligence algorithms, energy-efficient computational architectures, and emerging memory devices. As a Doctoral Fellow, he worked with Prof. Chetan Singh Thakur at the Indian Institute of Science and Prof. Shantanu Chakrabartty at Washington University in St. Louis, where he designed several novel architectures for machine learning and edge computing tasks that are truly inspired by the brain. Currently, his research focuses on addressing the overgrowing hardware bottleneck for artificial intelligence hardware through the analog route. The goal is to create a novel technological solution to cater to the needs of the present generation of students for designing ultra-low power, energy-efficient, and cost-effective analog accelerator chips for AI tasks that can be scaled across process technology nodes.
Further Activities to have a look at