Esteemed Professor Yiran Chen Sheds Light on the Future of AI on Edge Devices
31 Oct 2024
Professor Chen sheds light on utilising ‘Mix-Precision with Bit-Level Sparsity’ technique to enhance AI efficiency on edge devices.
The lecture attracts a full house of enthusiastic participants, sparking an engaging discussion on advancing AI in functionality, efficiency, and trustworthiness.
The enlightening lecture concludes with a group photo of Professor Chen (sixth from left) and faculty members.
Distinguished Professor Yiran Chen, a prominent figure in the field of AI from Duke University, captivated his audience with a thought-provoking lecture titled “Big AI for Small Devices” on 23 October 2024.
In this lecture, Professor Chen explored the significant challenges of deploying advanced AI models on resource-constrained edge devices, such as smartphones and IoT sensors. He emphasised the necessity of simplifying both the topology and operations of neural networks to enhance efficiency. To bridge the gap between large AI models and edge devices, he introduced “Least Absolute Shrinkage and Selection Operator” as an effective mechanism to minimise the size of neural networks as well as computational complexity. Furthermore, Professor Chen's insights into Mixed-Precision Quantization in deep neural networks not only led to significant improvements in AI model performance but also greatly reduced memory usage on edge devices, paving the way for future innovations in AI deployment.
Professor Yiran Chen holds the esteemed title of John Cocke Distinguished Professor of Electrical and Computer Engineering at Duke University. He serves as the principal investigator and director of the NSF AI Institute for Edge Computing - Athena, the NSF Industry-University Cooperative Research Center for Alternative Sustainable and Intelligent Computing (ASIC), and the co-director of the Duke Center for Computational Evolutionary Intelligence (DCEI).
