Future Directions in High-Dimensional Data Analysis: New Methodologies: New Data Types and New Applications

23 June to 27 June 2008

Isaac Newton Institute for Mathematical Sciences, Cambridge, UK

Organisers: Dr David Barber (University College London), Professor Iain Johnstone (Stanford University), Dr Richard Samworth (University of Cambridge) and Professor Michael Titterington (University of Glasgow)

in association with the Newton Institute programme Statistical Theory and Methods for Complex, High-Dimensional Data (7 January to 27 June 2008)

Programme | Participants | Contributed Talks | Application | Accommodation and Cost | Workshop Home Page

Accepted Posters:

Name Title Abstract Poster
Bochkina, N Bayesian clustering of high-dimensional data via a directional approach Abstract
Chung, Y Nonparametric bayes conditional distribution modeling with variable selection Abstract
Ingleby, B Analysis of large-scale data-sets in atmospheric data assimilation Abstract
Kaban, A The concentration of the L2 distance in latent variable models and some consequences Abstract
Lei, J On stability and sparsity of ensemble kalman filters Abstract
Li, Y Deciding the dimension of effective dimension reduction space for functional data Abstract
Lounici, K Sup-norm convergence rate and sign concentration property of Lasso and Dantzig estimators Abstract
Ma, Z Accuracy of the Tracy-Widom limit for extreme eigenvalues in white Wishart matrices Abstract
Nicodemus, KK Correlation between Predictors impacts machine learning variable importance measures: implications for genetic studies Abstract
Rothman, A Regularization of the Cholesky factor of the covariance matrix in high dimensions Abstract
Seker, H A novel feature extraction and selection methods in proteo-informatics for protein characterisation Abstract
Yao, Y Functional component pursuit for small n large p data Abstract
Yi, G Model selection for Gaussian process regression Abstract

Information for Conference Participants | Local Information | Newton Institute Map | Statistical Theory and Methods for Complex, High Dimensional Data | Workshops | Newton Institute Home Page