Hosted by the Interactive AI Centre for Doctoral Training and the Intelligent Systems Lab

Contact iai-cdt@bristol.ac.uk if you would like lunch, served between 13.30-14.00 

Abstract: AI-driven data centres, operating continuously and predominantly powered by fossil fuels, contribute significantly to global greenhouse gas emissions (2.5-3.7%). The widespread use of large foundation models such as ChatGPT exacerbates this environmental impact. This talk explores strategies for mitigating AI's carbon footprint through model acceleration, aiming to significantly reduce computations while maintaining the accurate of deep learning models. This session will spotlight various model acceleration techniques, including Neural Architecture Search, Knowledge Distillation, Quantisation, and many others. These techniques have successfully been applied to computer vision and large language models. Beyond academic advancements, the talk will also delve into successful industrial applications. Last, it will outline potential future research directions in the field of model acceleration.

Bio: Dr. Guosheng Hu is a senior lecturer of AI in University of Bristol. Before that, he served as the Head of Research at Oosto (a Unicorn of AI) from 2016 to 2024. Prior to his role at Oosto, he was a Research Fellow in the THOTH team at INRIA Grenoble Rhone-Alpes, France. Dr. Hu earned his PhD under the supervision of Prof. Josef Kittler at the University of Surrey, UK. His expertise lies in the intersection of computer vision and deep learning. With a robust academic background, he has published numerous research papers at major conferences and journals. His homepage is https://huguosheng.github.io/.