A View in Old and New Machine Learning

Explore the evolution of machine learning techniques in this comprehensive seminar by Professor Wray Buntine from Monash University. Delve into both traditional and modern approaches to machine learning, starting with hierarchical probabilistic models like k-dependence Bayesian networks and hierarchical topic models. Examine the state-of-the-art applications of deep neural networks in multi-task learning, document classification, and learning to learn. Compare the capabilities of old and new technologies, and consider their relevance in today's landscape. Discover efforts to combine traditional statistical methods with modern deep learning approaches, providing a holistic view of the field's progression and future directions.