Automatic experimentation refers to a sequential data-collection process where data are analysed and the next experiment to run is chosen and performed with little to no human involvement. The presentations in this session will showcase recent developments in this area motivated by real applications in science and engineering
Programme:
14:00 Welcome
14:05 Entropy-based adaptive design for contour finding and estimating reliability - Robert Gramacy (Virginia Tech, USA)
14:55 Statistical methods for automatic optimisation of flow chemistry reactions - Tim Waite (University of Manchester, UK)
15:45 Break
15:55 Optimal Bayesian Design of Finitely Sequential Experiments with Deep Deterministic Policy Gradient - Xun Huan (University of Michigan, USA)
16:45 Questions and comments
17:00 End
Speakers:
Tim Waite (University of Manchester, UK)
Xun Huan (University of Michigan, USA)
Robert Gramacy (Virginia Tech, USA)
Chair:
Antony Overstall (University of Southampton, UK)
Free to RSS Fellows
£10 for non-Fellows