Sumon Dey

Papers

Total Citations

H-Index

About

Sumon Dey is a researcher whose work sits at the intersection of computer architecture and artificial intelligence, with a particular focus on hardware acceleration for neural networks. His most-cited paper, "Processor-in-memory support for artificial neural networks" (2016, 6 citations), explores how processing-in-memory (PIM) architectures can be leveraged to accelerate artificial neural network (ANN) computations. This work addresses a critical challenge in modern computing: the need for low-power, real-time processing in applications like autonomous vehicles, robotics, and data mining. By proposing hardware-level support for ANNs within memory, Dey contributes to the broader effort of making deep learning more efficient and deployable in resource-constrained environments. While his citation count is modest, his research aligns with a growing interest in neuromorphic and in-memory computing, areas that are gaining traction as the limits of traditional von Neumann architectures become apparent. Dey’s work offers valuable insights for students and researchers exploring how to bridge the gap between algorithmic advances in AI and the physical constraints of hardware.