Title: Compression and Training Acceleration of Deep Learning Models
Time: June 5, 2024, 14:00-15:00
Venue: Room 453, Information Building
Tencent Meeting ID: 774-138-956
Abstract
Deep learning has achieved remarkable results in fields such as computer vision and natural language processing. However, as the performance of deep learning models improves, their computational complexity also increases, posing challenges to the efficiency of model training and inference. Under these conditions, model compression and acceleration have become crucial for the broader application and deployment of artificial intelligence. This report will briefly introduce the current challenges in the field of model compression and acceleration and present the latest research progress our team has made in lightweight model compression and training acceleration over the past few years.
Peisong Wang，Ph.D.，is an associate researcher at the Institute of Automation, Chinese Academy of Sciences, and a supervisor for master's students. He obtained his bachelor's degree from Shandong University in 2013 and his doctorate from the University of Chinese Academy of Sciences in 2018. His current research mainly focuses on efficient intelligent computing, neural network acceleration, and compression. He has published over 40 papers in international journals and conferences such as IEEE TPAMI, TNNLS, NeurIPS, ICML, CVPR, ICCV, and ECCV. He has led the National Natural Science Foundation of China's Youth Fund and the Strategic Priority Research Program of the Chinese Academy of Sciences, and has participated in national key research and development programs, key funds, as well as multiple research projects with companies including Huawei, Alibaba, and Samsung. He has been awarded the Nvidia Scholarship, the IEEE International Standard Outstanding Contribution Award, and won the MicroNet Challenge in the NeurIPS International Neural Network Compression Competition. He has also been selected as a specially-appointed backbone researcher at the Chinese Academy of Sciences, as well as for the Beijing Association for Science and Technology's Young Talent Support Program, the Microsoft Asia Research Institute's 'Star Cast Plan,' and the CCF-Baidu Sangfor Scholar, among others.
