As the driving force behind all the recent high profile advances in artificial intelligence (AI), machine learning represents one of the most important technologies of the modern era. But how did we get here? How did something that was so low-key as to be practically invisible to the general public as recently as 2013, make the popular headlines on almost a daily basis merely a decade later?
This talk seeks to address that question, charting the irresistible rise of machine learning over recent decades in the context of some 80 years of history. Adopting a tone suitable for the general audience, the talk will attempt to shed light on many of the prime causes, tease out some of the critical contributions and identify past (and possible future) crises. Does machine learning (in the guise of AI) really pose an existential threat, or does it offer the opportunity for unrivalled future prosperity, "all watched over by machines of loving grace"?
Mike Tipping is Professor of Machine Learning in the Department of Computer Science at the University of Bath. Before joining the University, Mike spent several years in industry: eight years at Microsoft Research, six years running an independent statistical consultancy and two years as Director of Science at Featurespace. He has also worked with a number of start-ups, including Babylon Healthcare, Ninety Percent of Everything, Homelink Technologies and Propflo. Combining interests in both theory and application of AI, Mike is known for the introduction of the "probabilistic PCA" data science framework, the invention of the "relevance vector machine" predictive model, the origination of the "Drivatar AI" concept behind Microsoft's "Forza Motorsport" Xbox franchise, and the development of an award-winning machine learning-based payment fraud prevention system.