AD Model Builder: using automatic differentiation for statistical inference of highly parameterized complex nonlinear models

By David A. Fournier, Hans Julius Skaug, John Ancheta, James Ianelli, Arni Magnusson, Mark N. Maunder, Anders Nielsen, John Sibert. Optim. Methods Softw., 2012

1,897 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/723bf9d10ec0a9727d476b5833a1eb244e205eca

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Control and Systems Engineering

Citations in the corpus

Long-term effects of an offshore wind farm in the North Sea on fish communities (2015, methodology): “…a negative binomial distribution (after comparison with a Poisson distribution) and were analysed by a general linear mixed effect model (GLMM) for discrete data using the R software environment (R Core Team 2011) and the AD Model Builder package glmmADMB (Fournier et al. 2012, Skaug et al. 2012).”
Testing the roles of competition, facilitation and stochasticity on community structure in a species‐rich assemblage (2014, methodology): “Equations 2 and 3 were fitted using ADMB (Fournier et al. 2012).”
Accounting for correlated observations in an age-based state-space stock assessment model (2016, background): “(2014) were of the second kind due to computational convenience when using the AD Model Builder software (Fournier et al., 2012), where the posterior distributions of Yi|N are readily available, but the one-step conditionals Yi|i21 are not.”
Measuring uncertainty in fisheries stock assessment: the delta method, bootstrap, and MCMC (2013, methodology): “Automatic differentiation (Griewank and Corliss 1991; Fournier et al. 2012) facilitates the use of the delta method with complex models, where derived quantities are not simple functions of estimated parameters, by applying automated algorithms to compute the partial derivatives.”
Solving Irregular and Data-enriched Differential Equations using Deep Neural Networks (2019, methodology): “can be readily and robustly evaluated using reverse mode automatic differentiation [25].”
A novel length-based empirical estimation method of spawning potential ratio (SPR), and tests of its performance, for small-scale, data-poor fisheries (2015, methodology): “To ensure that there was no possibility of crosscontamination of parameter values between the OM and the estimation model, the estimation model was coded separately in ADMB (Fournier et al., 2012).”