Suggested Topics

[mebrooks]

This could be a good place to make and preliminarily discuss suggested topics before they are added to the schedule.

[MarkMaunder]

Discuss relationships between CAPAM, the next generation general stock assessment program, ADMB/TMB, and the ICES global stock assessment initiative.

[skaug]

TMB instruction videos could be a topic this year also. To get anywhere
we would need the involvement of 4-5 people for at least two days.
I could act as the producer like last year. First step is a plenary discussion
about which topics to cover.

[mintoc]

Sorry if this has already happened - how about a full TMB training course (theory, implementation, example applications)? Not specifically linked to stock assessment (the ICES course does that excellently) but something more general to engage broader communities who would benefit from TMB.

[AaronGreenberg]

I would love to see a comparison between how autodiff and cppAD implement forward accumulation differentiation! It really seems that the ADMB/TMB project is based around developing c++ classes that can automatically compute the Jacobian of a function, but I don't really know how those libraries work and what the differences between them are.

[AaronGreenberg]

Obviously there are sources like this:
http://www.robots.ox.ac.uk/~tvg/publications/talks/autodiff.pdf
However, they do not about a particular C++ library. I think Stan also uses a c++ autodiff library?
It seems the better documented these libraries are the more likely someone might be tempted to contribute code to ADMB/TMB rather than write their own.

[mebrooks]

We should discuss potential working groups that could be funded through rOpenSci.