Talk title: Machine learning after the deep learning revolution

Speaker: Wray Buntine, VinUniversity (báo cáo trực tiếp)

Time: 15:30-17:00, Thursday, August 18, 2022.

Seminar: Hybrid seminar [Registration here]

Abstract: Deep learning has produced a huge leap in performance for natural language processing, image processing and other areas. The 2013 IEEE T.PAMI review by Bengio, Courville and Vincet provides perspective but also demonstrates their mastery of a broad range machine learning (ML) and artificial intelligence (AI) theories: Bayesian modelling, probabilistic graphical models, Markov models, causality, reinforcement learning, and so forth. While its well known to older machine learning researchers that the founders of deep learning are extremely knowledgable in the broader areas, newer converts are not trained this way.  This talk will discuss why deep learning has been a success, relating to prior methods, and present some recent techniques such as pre-training and knowledge transfer within this perspective.

Bio: Wray Buntine is a full professor and Director of Computer Science at VinUniversity in Hanoi from December 2021 after 8 years at Monash University where he Director of Machine Learning and founding director of the Master of Data Science.  He was previously at NICTA in Canberra, University of Helsinki, NASA Ames Research Center, University of California, Berkeley, and Google.  He is on several journal editorial boards and has been senior programme committee member for premier conferences such as IJCAI, UAI, ICLR, NeurIPS, ACML and SIGKDD.  He has over 200 academic publications, several software products and two patents.