BEGIN:VCALENDAR VERSION:2.0 PRODID:-//Memento EPFL// BEGIN:VEVENT SUMMARY:IC Colloquium : Scalable Gaussian Processes for Scientific Discove ry DTSTART:20160208T101500 DTEND:20160208T113000 DTSTAMP:20260407T100410Z UID:e7fe6e90205f261548ef9f3ccca6db11916e79962aa00e09f1841b54 CATEGORIES:Conferences - Seminars DESCRIPTION:By : Andrew Gordon Wilson - Carnegie Mellon University\nIC Fac ulty candidateAbstract :\nEvery minute of the day\, users share hundreds o f thousands of pictures\, videos\, tweets\, reviews\, and blog posts. More than ever before\, we have access to massive datasets in almost every are a of science and engineering\, including genomics\, robotics\, and astrono my.  These datasets provide unprecedented opportunities to automatically discover rich statistical structure\, from which we can derive new scienti fic discoveries. Gaussian processes are flexible distributions over functi ons\, which can learn interpretable structure through covariance kernels. In this talk\, I introduce a Gaussian process framework which is capable o f learning expressive kernel functions on massive datasets.  I will show how this framework generalizes a wide family of scalable machine learning approaches\, including deep learning models\, and allows one to exploit mo del structure for significant further gains in scalability and accuracy\, without requiring severe assumptions.  I will then discuss how we can use this framework for reverse engineering human learning biases\, crime pred iction\, modelling the impacts of vaccine introduction\, image inpainting\ , video extrapolation\, and discovering the structure and evolution of sta rs.Bio :\nAndrew Gordon Wilson is a Postdoctoral Research Fellow in the Ma chine Learning Department at Carnegie Mellon University working with Eric Xing and Alexander Smola.  Andrew received his PhD in machine learning fr om the University of Cambridge in 2014\, supervised by Zoubin Ghahramani.\ nAndrew's research interests include probabilistic machine learning\, scal able inference\, Gaussian processes\, kernel methods\, Bayesian modelling\ , nonparametrics\, and deep learning.  Andrew received the G-Research Out standing Dissertation Award in 2014 and the Best Student Paper Award at th e Conference on Uncertainty in Artificial Intelligence in 2011.More inform ation LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420 STATUS:CONFIRMED END:VEVENT END:VCALENDAR