Discovering Structure in Data with Bayesian Nonparametrics

A perpetual challenge in statistical modelling is trying to find the parsimonious complexity in the data. That is, balancing simplicity in our explanations of the world with the flexibility that is required to capture the rich variation that occurs in real data. One remarkable class of mathematical tools for balancing these extremes are Bayesian nonparametric model, which enable one to specify an infinite-dimensional model, while still manipulating it tractably on a finite computer. Such models mean that our explanations for the world can grow in complexity precisely to the extent that the data allow it.
Graph-Sparse LDA: A Topic Model with Structured Sparsity. Doshi-Velez, F, Wallace B, Adams RP.  Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence. Submitted.   { arXiv:1410.4510 [stat.ML] | PDF }
Gaussian Process Kernels for Pattern Discovery and Extrapolation. Wilson, AG, Adams RP.  Proceedings of the 30th International Conference on Machine Learning. 2013.   { arXiv:1302.4245 [stat.ML] | PDF | Code }
Archipelago: Nonparametric Bayesian Semi-Supervised Learning. Adams, RP, Ghahramani Z.  Proceedings of the 26th International Conference on Machine Learning. 2009.   { PDF }