Biostatistics

I am interested in applications of statistics and probability to biology, especially to marine ecology. I have developed methods for line transect surveys and CKMR (close-kin mark recapture) methods in particular. Key reference: Bravington, MV, Skaug, HJ, Anderson, EC (2016). Close-kin Mark-Recapture. Statistical Science.

Computational statistics

My opinion for a long time has been that Statisticians should have Automatic Differentiation (AD) in their toolbox. My conjecture has been that a combination of AD and the Laplace approximation is a powerful platform for fitting models with continuous latent variables. Software projects in which this idea is implemented are:

• TMB: https://github.com/kaskr/adcomp

• ADMB: http://admb-project.org/

Artificial intelligence (AI)

Witnessing the recent breakthroughs in the field, my research interests have turned towards AI. Techniques such as variational autoencoders and diffusion models are probabilistic in nature, and Backpropagation in Deep Learning is the same computational technique as AD, which I have been working with for 20 years.

Editorial services

co-Editor in chief of Scandinavian Journal of Statistics (2018–2021)

Teaching at UiB (last 5 years)

• Spring: STAT110 Basic course in statistics
• Fall:  STAT110 Basic course in statistics

Short courses / workshops (in the past)

• CKMR (Close-kin mark recapture) course (co-thought) Dalhousie University, Halifax, July 2019.
• TMB (Template Model Builder) course (co-thought) at IMR Bergen, June 2019.
• Modelling in TMB (Template Model Builder). Course (co-thought) at IMR Bergen,  June 2018.
• An introduction to fish stock assessment in TMB (Template Model Builder). Course given Oct. 2017 in La Jolla

Google scholar profile 

From CRIStin (Current research information system in Norway):