M
Matthew M. Dunlop
Researcher at California Institute of Technology
Publications - 26
Citations - 537
Matthew M. Dunlop is an academic researcher from California Institute of Technology. The author has contributed to research in topics: Bayesian probability & Markov chain Monte Carlo. The author has an hindex of 11, co-authored 25 publications receiving 434 citations. Previous affiliations of Matthew M. Dunlop include New York University.
Papers
More filters
Journal ArticleDOI
Hierarchical Bayesian level set inversion
TL;DR: This paper demonstrates how the scale-sensitivity of the Bayesian approach can be circumvented by means of a hierarchical approach, using a single scalar parameter, leading to well-defined Gibbs-based MCMC methods found by alternating Metropolis–Hastings updates of the level set function and the hierarchical parameter.
Journal ArticleDOI
How Deep Are Deep Gaussian Processes
TL;DR: In this article, the potential utility of deep Gaussian processes has been shown, where deep structures are probability distributions, designed through hierarchical construction, which are conditionally conditionally optimal.
Journal ArticleDOI
Large data and zero noise limits of graph-based semi-supervised learning algorithms
TL;DR: Scalings in which the graph Laplacian approaches a differential operator in the large graph limit are used to develop understanding of a number of algorithms for semi-supervised learning; in particular the extension, to this graph setting, of the probit algorithm, level set and kriging methods are studied.
Journal ArticleDOI
Iterative Updating of Model Error for Bayesian Inversion
TL;DR: An algorithm which iterates this idea to update the distribution of the model error, leading to a sequence of posterior distributions that are demonstrated empirically to capture the underlying truth with increasing accuracy, and it is proved that each element of the sequence converges in the large particle limit.
Posted Content
The Bayesian Formulation of EIT: Analysis and Algorithms
TL;DR: In this article, a rigorous Bayesian formulation of the EIT problem in an infinite dimensional setting was provided, leading to well-posedness in the Hellinger metric with respect to the data.