[Turing-Southampton] S3RI seminar: Claire Gormley, Thursday 2-3pm

[Helen Ogden]

Dear all,

On Thursday (25 October) at 2pm in 54 / 7035 (7B), we have an S3RI 
seminar from Claire Gormley (University College Dublin) on "Infinite 
Mixtures of Infinite Factor Analysers". Details are given below.

The seminar will also be available via a live web-cast at
https://coursecast.soton.ac.uk/Panopto/Pages/Viewer.aspx?id=8cccbf5a-af48-4e7d-a91f-4233a3189d03 
<https://www.google.com/url?q=https%3A%2F%2Fcoursecast.soton.ac.uk%2FPanopto%2FPages%2FViewer.aspx%3Fid%3D8cccbf5a-af48-4e7d-a91f-4233a3189d03&sa=D&usd=2&usg=AFQjCNHmWfm6CiNrusYdIiS4JgbDp6qSbA>

The talk will be followed by tea and cake in the staff reading room on 
level 4 of building 54.

All are welcome!

Best wishes,

Helen

Infinite Mixtures of Infinite Factor Analysers (IMIFA)

Claire Gormley, University College Dublin

Factor-analytic Gaussian mixture models are often employed as a 
model-based approach to clustering high-dimensional data. Typically, the 
numbers of clusters and latent factors must be specified in advance of 
model fitting, and the optimal pair selected using a model choice 
criterion. For computational reasons, models in which the number of 
latent factors is common across clusters are generally considered.

Here the infinite mixture of infinite factor analysers (IMIFA) model is 
introduced. IMIFA employs a Poisson-Dirichlet process prior to 
facilitate automatic inference on the number of clusters. Further, IMIFA 
employs shrinkage priors to allow cluster specific numbers of factors, 
automatically inferred via an adaptive Gibbs sampler. IMIFA is presented 
as the flagship of a family of factor-analytic mixture models, providing 
flexible approaches to clustering high-dimensional data.

Applications to benchmark and real data sets illustrate the IMIFA model 
and its advantageous features: IMIFA obviates the need for model 
selection criteria, reduces model search and associated computational 
burden, improves clustering performance by allowing cluster-specific 
numbers of factors, and quantifies uncertainty in the numbers of 
clusters and cluster-specific factors. The IMIFA R package, available on 
CRAN, facilitates implementation of our method.

This is joint work with Keefe Murphy (University College Dublin) and 
Cinzia Viroli (Universita di Bologna)

For the current schedule of S3RI seminars, see 
https://tinyurl.com/s3riseminar

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ecs.soton.ac.uk/pipermail/turing-southampton/attachments/20181022/b570faab/attachment.html