Mehdi Ghasemi

About

Mehdi Ghasemi is a leading researcher at the intersection of embedded systems and machine learning, with a focus on optimizing deep neural network (DNN) deployment on edge multiprocessor systems-on-chips (MPSoCs). His most cited work, "CAMDNN: Content-Aware Mapping of a Network of Deep Neural Networks on Edge MPSoCs" (2022, 5 citations), addresses the critical challenge of efficiently executing heterogeneous ML workloads at the edge. Ghasemi’s key contribution lies in developing content-aware mapping strategies that maximize resource utilization across diverse DNN models, accounting for both model and system heterogeneity. This work is pivotal for enabling real-time, energy-efficient inference in resource-constrained edge devices, such as those used in autonomous systems and IoT. His research bridges the gap between algorithmic ML advances and practical hardware constraints, offering scalable solutions for deploying complex neural networks in the field. By tackling the bottleneck of efficient inference execution, Ghasemi’s contributions have significant implications for advancing edge AI, making him a notable figure in the domain of embedded deep learning.

Research Focus

Key Achievements

Top Papers

1. 1
   CAMDNN: Content-Aware Mapping of a Network of Deep Neural Networks on Edge MPSoCs

   5 citations · 2022

Key Collaborators

Contact & Links