Data-efficient Learning with Dataset Condensation

[CFAR Rising Star Lecture Series]
Data-efficient Learning with Dataset Condensation by Dr Zhao Bo
23 Mar 2023 | 4.00pm (Singapore Time)

In numerous fields, larger datasets are increasingly required to achieve state-of-the-art performance. Storing these datasets and training models on them has become significantly more expensive, especially when validating multiple model designs and hyper-parameters. To address this, Dr Zhao Bo and his team have proposed Dataset Condensation, a training set synthesis technique for data-efficient learning to condense a large dataset into a small set of informative synthetic samples for training deep neural networks from scratch. 

In this talk, Dr Zhao will present his team’s recent progress in dataset condensation research and demonstrate some promising results from the application of dataset condensation to continual learning, neural architecture search and privacy-preserving learning.

Dr Zhao Bo
Beijing Academy of Artificial Intelligence

Dr Zhao Bo is a researcher in the Beijing Academy of Artificial Intelligence (BAAI). He received his Ph.D. from the University of Edinburgh, under the supervision of Dr. Hakan Bilen (Associate Professor), and M.Eng. from Peking University, where he worked under Prof. Wang Yizhou. He is currently working on Data-centric AI, Data-efficient Learning and Dataset Condensation. Dr Zhao is also the recipient of the International Conference on Machine Learning (ICML) 2022 Outstanding Paper Award.