Ajay Jasra - Unbiased estimation of the gradient of the log-likelihood for partially observed diffusions
|Starts:||12:00 24 Sep 2019|
|Ends:||13:00 24 Sep 2019|
|What is it:||Seminar|
|Organiser:||Department of Mathematics|
|Who is it for:||University staff, External researchers, Current University students|
|Speaker:||Professor Ajay Jasra|
Join us for this research seminar, part of the SQUIDS (Statistics, quantification of uncertainty, inverse problems and data science) seminar series.
Abstract: In this work we present a new coupled simulation method for the unbiased estimation of the gradient of the log-likelihood for a class of partially observed diffusion processes. This is of interest in stochastic gradient algorithms, which typically require such unbiasedness. Intrinsically, given only access to standard time-discretizations of diffusion processes (such as Euler), we present a methodology which provides unbiased estimates of gradient of the log-likelihood which has no time-discretization error. This method also provides estimates with finite variance and expected cost. Some preliminary simulation results are also given. This is a joint work with Jeremy Heng (ESSEC Singapore) and Jeremie Houssineau (Warwick).
Professor Ajay Jasra
Role: Professor of Applied Mathematics and Computational Science
Organisation: King Abdullah University of Science and Technology
Biography: Ajay Jasra is currently professor of applied mathematics and computational science at the Computer, Electrical and Mathematical Sciences and Engineering Division at King Abdullah University of Science and Technology. Before that he was Deans' Chair Associate Professor at the National University of Singapore and Assistant Professor (Lecturer) at Imperial College London. Prof Jasra works on the interface of Bayesian statistics, Monte Carlo Computation and inverse problems, with several application areas in biological and financial modleling.
Travel and Contact Information
Frank Adams 1
Alan Turing Building