Dennis Prangle

Dennis Prangle is an Associate Professor in Statistics at the University of Bristol, conducting research at the intersection of Bayesian statistics and machine learning. His academic profile demonstrates expertise in developing novel computational inference methods with applications across multiple scientific domains.

Dr. Prangle's primary research interests include:

• Approximate inference methods such as simulation-based inference and variational techniques
• Likelihood-free inference through Approximate Bayesian Computation (ABC)
• Experimental design for high-dimensional problems
• Applications in population genetics, physics, ecology, and epidemiology
• Stochastic differential equations and composite likelihood approaches

His publication record shows consistent methodological contributions with increasing integration of machine learning techniques. Recent work focuses on normalizing flows with flexible tails for improved density estimation, Bayesian emulation of complex systems, and optimal combination of composite likelihoods. His research bridges theoretical statistics with practical applications in financial modeling, infrastructure engineering, and fair classification algorithms.

Dr. Prangle maintains an active academic blog where he discusses technical aspects of Bayesian statistics, experimental design, and computational methods, demonstrating his commitment to scholarly communication. His detailed posts on topics like Fisher information gain versus Shannon information gain in experimental design highlight his theoretical contributions to the field.