Tien-Ju Yang

Papers

Total Citations

4,051

H-Index

About

Tien-Ju Yang is a researcher specializing in the efficient deployment of deep neural networks (DNNs) and edge computing, with a focus on making artificial intelligence accessible on resource-constrained hardware. His work sits at the intersection of machine learning, computer architecture, and embedded systems — areas of increasing importance as AI moves beyond data centers into real-world devices. Yang's most influential contribution is his co-authorship of "Efficient Processing of Deep Neural Networks: A Tutorial and Survey" (2017), a foundational reference in the field that has accumulated nearly 4,000 citations, making it one of the most widely read works in AI hardware research. The paper comprehensively addresses the computational challenges of DNNs and surveys techniques for improving their efficiency — an essential resource for students and practitioners alike. His work on **FastDepth** (2019) further demonstrates his commitment to practical AI deployment, presenting a method for real-time monocular depth estimation optimized for embedded systems — a capability critical for robotics applications including mapping, localization, and obstacle avoidance. Overall, Yang's research has meaningfully shaped how the community thinks about bridging the gap between powerful AI models and the hardware constraints of the real world.