Navigation

Researchers at UCD

researcher

Michael Salter-Townshend

Lecturer/Assistant Professor

School Of Mathematics & Statistics
Science Centre - South
Belfield
Dublin 4

Tel: +353 1 7162487
Email: michael.salter-townshend@ucd.ie

Biography

I am an Assistant Professor of Statistics in University College Dublin since Jan 2017. 

My primary research interest is in population genetics, with a focus on human genomes. This involves modelling the patterns of shared variation between individuals and can shed light on questions of human demographic history, natural selection, and disease association mapping. My particular focus is on building sophisticated models for admixture; this 
is when two or more previously diverged sub-populations come together and it leads to interesting structure in the DNA of ancestors of the mixing groups. To this end I'm building extensions to some existing Hidden Markov Models for population genetics.

I'm also active in several other areas of applied computational statistics, including social network analysis and palaeoclimate reconstruction. See my publications for more information. 

I spent 3 years as a postdoc working with Prof. Simon Myers in the University of Oxford where I was also a Junior Research Fellow at Mansfield College from 2014 to 2016. 

Before this, I was a postoc with Prof. Brendan Murphy in UCD, Dublin on a network analysis project as part of Clique. Clique was an SFI funded multidisciplinary research cluster on network analysis. I worked on developing computationally efficient models for various types of network model. I developed Variational Bayesian methods and also used MCMC and EM algorithms.

My PhD was with Prof. John Haslett in Trinity College Dublin on a palaeoclimate reconstruction project. During my PhD I spent time working with Havard Rue in NTNU, Trondheim developing algorithms for Integrated Nested Laplace Approximations to the hierarchical Bayesian model used for palaeoclimate reconstruction. 

Professional

Honours and Awards

Year: 2014.
Title: Junior Research Fellowship at Mansfield College, University of Oxford
       

Conference Contributions

Michael Salter-Townshend, Simon Myers (2016) Multiway Admixture Event Modelling without Specifying Groups. [Oral Presentation], PROBABILISTIC MODELING IN GENOMICS, University of Oxford , 12-SEP-16 - 14-SEP-16.
Thomas Brendan Murphy and Michael Salter-Townshend; (2009) Variational Bayesian Inference for the Latent Position Cluster Model. [Chaired Session], Statistical Methods for the Analysis of Network Data in Practice, UCD , 15-JUN-09 - 17-JUN-09.
Michael Salter-Townshend; (2010) Variational Bayesian Inference for the Latent Position Cluster Model. [Chaired Session], Workshop on Modeling and Computation for Social Networks, Whistler, BC, Canada , 26-JUN-10 - 27-JUN-10.
Michael Salter-Townshend; (2007) Gaussian Approximation Techniques. [Poster Presentation], Conference of Applied Statistics Ireland, Dundrum, Tipperary , 15-JUN-07 - 16-JUN-07.
Michael Salter-Townshend; (2006) Fast Approximations to High Dimensional Response Surfaces. [Oral Presentation], High Performance Computing in Statistical Inference, Trinity College Dublin , 17-AUG-06 - 17-AUG-06.
Michael Salter-Townshend; (2006) Zero-Inflation of Compositional Data. [Oral Presentation], 21 st International Workshop on Statistical Modelling, NUI Galway , 19-JUL-06 - 19-JUL-06.
Michael Salter-Townshend; (2006) Additive Modelling of Response Surfaces. [Poster Presentation], Conference of Applied Statistics in Ireland, Tralee , 17-JUN-06 - 19-JUN-06.
Michael Salter-Townshend; (2006) Zero-Inflation of Compositional Data. [Oral Presentation], International Conference of the Royal Statistical Society, Queen's University Belfast , 30-SEP-06 - 30-SEP-06.
Michael Salter-Townshend; (2006) Additive Modelling of Response Surfaces. [Oral Presentation], Young Statisticians Meeting, Edinburgh , 12-FEB-06 - 12-FEB-06.
Michael Salter-Townshend; (2005) Temporal Smoothing. [Poster Presentation], Conference of Applied Statistics in Ireland, Enniskillen , 18-JUN-05 - 20-JUN-05.

Committees

Committee : The Network seminar series, Dept. of Statistics, University of Oxford
Committee : Green Impact Team, Dept. of Statistics, University of Oxford
Committee : www.irishsupercomputerlist.org/
 

Education

Year 2009 Institution: Trinity College Dublin
Qualification: PhD Subject:
Year 2004 Institution: Trinity College Dublin
Qualification: MSc Subject: High Performance Computing
Year 2003 Institution: Trinity College Dublin
Qualification: BA (Mod) Subject: Computational Physics
       

Other Activities

Referee: 
Annals of Applied Statistics.
Journal of Statistics and Computing.
Network Science.
Journal of the Environmetrics Society.
Pattern Analysis and Applications.
International Statistical Institute World Congress 2011.
AISTATS 2012, 2013.
NIPS 2014. 

Publications

 

Book Chapters

Salter-Townshend, M. and Murphy, T.B. (2013) 'Sentiment Analysis of Online Media' In: Lausen, B., van del Poel, D. and Ultsch, A (eds). Algorithms from and for Nature and Life. Heidelberg: Springer. , pp.137-146 [DOI] Link to full text [Details]
 

Peer Reviewed Journals

Michael Salter-Townshend and Tyler H. McCormick (2017) 'Latent space models for multiview network data'. Annals of Applied Statistics, 11 (3):1217-1244. Available Online [DOI] [Details]
Qin, Xiangju and Cunningham, P\'adraig and Salter-Townshend, Michael (2016) 'Online Trans-dimensional von Mises-Fisher Mixture Models for User Profiles'. Journal of Machine Learning Research, 17 (200):1-51. [Details]
Qin, Xiangju and Cunningham, P\'adraig and Salter-Townshend, Michael (2015) 'The influence of network structures of Wikipedia discussion pages on the efficiency of WikiProjects'. Social Networks, 43 :1-15. [Details]
Salter-Townshend, M,Murphy, TB (2013) 'Variational Bayesian inference for the Latent Position Cluster Model for network data'. Computational Statistics and Data Analysis, 57 :661-671. [DOI] [Details]
Salter-Townshend, M., White, A., Gollini, I. and Murphy, T.B. (2012) 'Review of Statistical Network Modelling: Models, Algorithms and Software'. Statistical Analysis and Data Mining, 5 (4):243-264. [DOI] [Details]
Salter-Townshend, Michael and Murphy, Thomas Brendan (2014) 'Mixtures of biased sentiment analysers'. Advances in Data Analysis and Classification, 8 (1):85-103. [Details]
Salter-Townshend, Michael and Brendan Murphy, Thomas (2015) 'Role analysis in networks using mixtures of exponential random graph models'. Journal of Computational and Graphical Statistics, 24 (2):520-538. [Details]
Salter-Townshend, M,Haslett, J (2012) 'Fast inversion of a flexible regression model for multivariate pollen counts data'. Environmetrics, 23 :595-605. [DOI] [Details]
J. Haslett, M. Whiley, S. Bhattacharya, M. Salter-Townshend, S. Wilson, J.R.M Allen, B. Huntley, F.J.G. Mitchell; (2006) 'Bayesian Palaeoclimate Reconstruction'. Journal of the Royal Statistical Society. Series A, (Statistics in Society), 169 (3):1-36. [Details]
Parnell, Andrew C and Sweeney, James and Doan, Thinh K and Salter-Townshend, Michael and Allen, Judy RM and Huntley, Brian and Haslett, John (2015) 'Bayesian inference for palaeoclimate with time uncertainty and stochastic volatility'. Journal of the Royal Statistical Society. Series C: Applied Statistics, 64 (1):115-138. [Details]
M. Salter-Townshend, J. Haslett, N. Friel (2009) 'Discussion of `Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations' by Rue, Martino and Chopin'. Journal of the Royal Statistical Society, Series B, 71 . [Details]
 

Conference Publications

Salter-Townshend, M. and Murphy, TB.; (2009) Variational Bayesian inference for the latent position cluster model NIPS Workshop on Analyzing Networks and Learning with Graphs Available Online [Details]
Qin, X., Salter-Townshend, M., Cunningham, P. (2014) Exploring the Relationship between Membership Turnover and Productivity in Online Communities Eighth International Conference on Weblogs and Social Media (ICWSM 2014) [Details]
Salter-Townshend, M. and Murphy, T.B.; (2011) Sentiment Analysis of Online Media Proceedings of the 58th Congress of the International Statistical Institute Dublin, Ireland, Available Online [Details]
Salter-Townshend, M. and Murphy, T.B.; (2012) Sentiment Analysis of Online Media . In: Lausen, B., van del Poel, D. and Ultsch, A eds. Joint Conference of the German Classification Society (GfKl) and the German Association for Pattern Recognition (DAGM) [Details]
Michael Salter-Townshend and Anthony Brew; (2010) A Latent Space Mapping for Link Prediction Networks Across Disciplines: Theory and Applications [Details]
Thomas Brendan Murphy and Michael Salter-Townshend; (2009) Variational Bayesian Inference for the Latent Position Cluster Model Analyzing Networks and Learning With Graphs; NIPS [Details]
J. Haslett, A. Parnell and M. Salter-Townshend; (2006) Modelling Temporal Uncertainty in Palaeoclimate Reconstructions International Workshop on Statistical Modelling [Details]
M. Salter-Townshend and J. Haslett; (2006) Zero-Inflation of Compositional Data International Workshop on Statistical Modelling [Details]
                           

Dissertations/Theses

Michael Salter-Townshend; (2009) Fast Approximate Inverse Bayesian Inference in non-parametric Multivariate Regression (with application to palaeoclimate reconstruction). Dublin: Dissertations/Theses Available Online [Details]
                                   

Media

Salter-Townshend, M (2012) Analysing My Facebook Friends. Media Available Online [DOI] [Details]
                                                                   

Research

Research Interests

Population Genetics:
- Admixture models; Hidden Markov Models.
- Ancient DNA.  
- Summaries and tests for population structure. 
- Scalability to large datasets / sequencing data. 
 
Statistical and probabilistic modelling of social networks:
- Fast approximate inference routines for Bayesian latent variable network models
- Multiple-relations / multiview / multilevel network modelling.
- Community finding and block modelling.
- Link and covariate prediction.
- Role analysis of nodes in a network via ego-network modelling.
 
Inference Methodology:
- Variational Bayes.
- MCMC.
- Integrated Nested Laplace Approximations (INLA).
- EM algorithm.
- Hidden Markov Models.  
 
Sentiment Analysis:
- Modelling multiple imperfect annotations.
- Model Based Clustering of annotators.

Palaeoclimate Reconstruction:
- Modelling multivariate, zero-inflated, constrained, spatial counts data.
- Solution of Inverse Problems via inversion of Bayesian Hierarchical Models.
   

Current Postgraduate Students

JINBO ZHAO, Doctor of Philosophy (PhD)   -   Thesis Supervisor
Chiara Cotroneo, Doctor of Philosophy (PhD)   -   Thesis Co-Supervisor

Teaching

Teaching Philosophy

 

Enhancement of Teaching

Review and Enhancement of Curriculum

Developing as a Teacher

Innovation & Leadership

Collaborators

Internal Collaborators

Prof. Brendan Murphy

External Collaborators

Prof. Simon Myers
Prof. John Haslett