Yudong Tao
Papers
1
Total Citations
27
H-Index
1
About
Yudong Tao is a researcher at the forefront of energy-efficient deep learning and edge computing, with a primary focus on optimizing Deep Neural Network (DNN) training for resource-constrained hardware. His most cited work, "Challenges in Energy-Efficient Deep Neural Network Training with FPGA" (2020, 27 citations), critically examines the pressing need to deploy DNNs on edge devices—such as mobile phones, drones, and wearable technology—for real-time visual data processing. Tao’s major contribution lies in identifying and addressing the unique bottlenecks of local DNN training on FPGAs, a domain traditionally dominated by inference tasks. By highlighting strategies to reduce energy consumption without sacrificing accuracy, his research paves the way for autonomous, privacy-preserving AI systems that learn on the fly. This work has been instrumental for engineers designing next-generation edge AI hardware. Beyond this paper, Tao’s broader research portfolio explores efficient neural network architectures and hardware-software co-design, earning him recognition as a key voice in sustainable AI. His insights are particularly valuable for students and researchers seeking to bridge the gap between algorithmic performance and real-world deployment in energy-sensitive environments.
Research Focus
Key Achievements
Top Papers
- 1Challenges in Energy-Efficient Deep Neural Network Training with FPGA27 citations · 2020