Paul Keir

Papers

Total Citations

H-Index

About

Paul Keir is a researcher at the forefront of portable hardware acceleration and edge computing, with a focus on making high-performance computing accessible on resource-constrained devices. His most-cited work, "Porting SYCL accelerated neural network frameworks to edge devices" (2023), addresses the critical challenge of deploying neural networks on edge hardware without sacrificing performance. By leveraging the SYCL standard—a cross-platform abstraction layer for heterogeneous computing—Keir demonstrates how machine learning frameworks can be efficiently ported to edge devices, enabling real-time AI inference in applications like IoT, autonomous systems, and smart sensors. This contribution is particularly timely as the demand for distributed, low-latency processing grows. With 3 citations and growing interest, his work bridges the gap between cloud-scale AI and edge deployment, offering practical solutions for researchers and engineers. Keir’s research underscores the importance of software portability in an increasingly heterogeneous hardware landscape, and his findings provide a roadmap for optimizing neural network workloads on limited-power platforms. His efforts are shaping the future of edge intelligence, making him a key voice in the evolution of portable, accelerated computing.