Machine learning, or the study of algorithms that can learn and act, allows automated decision-making that is both scalable and free of human error. It is becoming increasingly apparent that many tasks and even jobs traditionally done by humans can be carried out in a fraction of the time and at a fraction of the cost by machines. Dr Michael Osborne, Associate Professor in Machine Learning, and Co-Director of the Oxford Martin Programme on Technology and Employment, will look at current advances in machine learning, and consider the applications these could have on future technologies.
Join in on Twitter #tomorrowtech
About the speaker
Michael Osborne is Co-Director on the Oxford Martin Programme on Economic and Technological Change, Associate Professor in Machine Learning, an Official Fellow of Exeter College, and a Faculty Member of the Oxford-Man Institute for Quantitative Finance, all at the University of Oxford.
His work centres on the design of intelligent algorithms capable of making sense of complex data. Building such algorithms requires the use of techniques from Machine Learning and Computational Statistics; he works to create modular numerical algorithms that speak the common language of probability theory. His work in non-parametric data analytics has been successfully applied in diverse and challenging contexts. These contexts range from astrostatistics, where his probabilistic algorithms have aided the detection of planets in distant solar systems, to zoology, where his work has helped to clarify how pigeons are able to navigate such extraordinary distances. He has particular expertise in active learning, Gaussian processes, changepoint detection, Bayesian optimisation and Bayesian quadrature.
Michael Osborne's career has been shaped by extensive engagement with industry, both in research and consultancy arrangements. His DPhil work on sensor networks had significant influence on the EPSRC/Industry research project ALADDIN, winner of the Engineer Award Aerospace and Defence 2009. His work on fault detection and big data analytics has been demonstrated within a variety of industrial contexts.
Most recently, he has addressed the broader societal consequences of machine learning and robotics. In particular, he's worked to analyse how intelligent algorithms might soon substitute for human workers, and to predict the resulting impact on employment. Most notably in the widely cited 2013 paper The Future of Employment: How susceptible are jobs to computerisation?
This latter work has enjoyed broad media coverage (featured in The Economist, the Financial Times, the Wall Street Journal, The Independent, ITV News and the BBC World Service) and has substantial policy implications related to the future of employment.