Sumit Diware

Papers

Total Citations

H-Index

About

Sumit Diware is a leading researcher at the forefront of energy-efficient computing, specializing in memristor-based architectures for edge-AI applications. His work directly addresses the critical challenge of bringing intelligent computation to resource-constrained IoT devices, where traditional von Neumann architectures fall short due to high power demands. Diware’s most cited paper, “Low-Power Memristor-Based Computing for Edge-AI Applications” (2021, 34 citations), proposes a novel computing paradigm that leverages memristors—non-volatile memory devices that can both store and process data—to drastically reduce energy consumption. This contribution is pivotal for enabling real-time AI inference on smart edge-devices, from personalized healthcare monitors to autonomous robotics, without relying on cloud connectivity. By demonstrating how memristor crossbar arrays can perform analog in-memory computing, Diware has helped pave the way for ultra-low-power neural network accelerators. His work stands out for its practical focus on bridging the gap between emerging nanodevice physics and deployable edge-AI systems, making him a key voice in the push toward sustainable, decentralized intelligence.