A tutorial on optimization, model fitting, and model comparison for neuroscience
Luigi Acerbi (International Brain Laboratory)
Geographical Sciences Building G.11N SR1
Hosted by the Neural Dynamics Forum.
In terms of topics, I will give a brief introduction to likelihood-based modeling and then focus on model fitting by maximum-likelihood, with a quick overview of optimization methods including state-of-the-art Bayesian optimization and BADS . I will then talk briefly about model selection metrics, pros and cons. Depending on the audience and if we have time, I will also touch upon more advanced topics, such as Bayesian posterior estimation and approximate inference techniques, such as VBMC  (on which I am giving a talk at 2pm on the same day at Queens building). The tutorial will be in Matlab, but many of the concepts (if not directly the techniques) are language-free and can be easily applied in your favourite language, be it Python, Julia, R, or BASIC.
 BADS repo: https://github.com/lacerbi/bads
 VBMC repo: https://github.com/lacerbi/vbmc
More can be found here: http://luigiacerbi.com/tutorials/
For further information contact Anne-Lenne Sax.