Joseph Doyle

About

Joseph Doyle is a researcher at the forefront of efficient machine learning for resource-constrained embedded systems. His primary research areas include binary neural networks, tiny machine learning (TinyML), and edge computing for industrial and sensory applications. Doyle's major contribution lies in pioneering memory-efficient deployment of binarized convolutional neural networks on microcontrollers—the ubiquitous platform for field applications. His most cited work, "Memory Efficient Binary Convolutional Neural Networks on Microcontrollers" (2022, 10 citations), demonstrates how binarization optimizes limited-resource devices, enabling advanced AI on low-power hardware. He further advanced this field with "A Tiny CNN for Embedded Electronic Skin Systems" (2022, 5 citations), showcasing compact neural architectures for tactile sensing. Together, these contributions address the critical challenge of bringing deep learning to the industrial edge, where memory and computational budgets are extremely tight. Doyle's work is particularly notable for bridging the gap between theoretical optimization and practical deployment, making him a key figure in the growing TinyML community.

Research Focus

Key Achievements

Top Papers

1. 1
   Memory Efficient Binary Convolutional Neural Networks on Microcontrollers

   10 citations · 2022
2. 2
   A Tiny CNN for Embedded Electronic Skin Systems

   5 citations · 2022

Key Collaborators

Contact & Links