Neil Thacker - Fundamental Issues Regarding Uncertainties in Artificial Neural Networks
|Starts:||14:00 17 Feb 2020|
|Ends:||15:00 17 Feb 2020|
|What is it:||Seminar|
|Organiser:||Department of Mathematics|
|Who is it for:||University staff, External researchers, Current University students|
(Zochonis Lecture Theatre D)
Join us for this seminar by Neil Thacker (Manchester) as part of the Mathematics in the Life Sciences Series.
The problem of assessment of uncertainty in non-linear systems is re-visited and used to motivate an approach for estimation of parameter uncertainty based upon a Bayesian/Frequentist synthesis. The method is applied to Artificial Neural Networks trained on a real-world data set for clinical decision support (dementing diseases). This example identifies fundamental issues regarding the general assessment of uncertainty for popular multi-variate methods such as Deep Learning.
To subscribe to the mailing list for this event series, please send an e-mail with the phrase “subscribe math-lifesci-seminar” in the message body to firstname.lastname@example.org
Role: Senior Lecturer
Organisation: University of Manchester
Travel and Contact Information
Lecture Theatre D