[Seminar] "Bayesian Inference and Experimental Design for Implicit Models" by Dr. Michael Gutmann

Dear all,

Neural Computation Unit (Doya Unit) would like to invite you to a seminar as follows.

Date: Friday, April 19, 2019
Time: 14:00 – 15:00
Venue: Meeting room D015, Lab1 Bldg.

Speaker: Dr Michael Gutmann
               School of Informatics, University of Edinburgh

Title:
Bayesian Inference and Experimental Design for Implicit Models
 
Abstract:
Bayesian experimental design involves the optimal allocation of resources in an experiment, with the aim of optimising cost and performance. For implicit (simulator-based) models, where the likelihood is intractable but simulating data from the model is possible, this task is particularly difficult and therefore largely unexplored. This is mainly due to technical difficulties associated with approximating posterior distributions and utility functions. In this talk, I present some work of my group on efficiently learning the posterior and on leveraging those results for experimental design.

Main reference:
 https://arxiv.org/abs/1810.09912 (AISTATS 2019, joint work with Steven Kleinegesse)

We hope to see many of you at the seminar.

Sincerely,
Neural Computation Unit
Contact: ncus@oist.jp