Using Bayesian Adaptive Stimulus Selection to Estimate Generalization Curves

dc.contributor.advisorSelen, Luc
dc.contributor.advisorHinne, Max
dc.contributor.authorNeacsu, Mihaela
dc.description.abstractIn my research, I implement the Bayesian Adaptive Stimulus Selection algorithm proposed by Kontsevich and Tyler (Kontsevich & Tyler, 1999) and adapt it in order to estimate the parameters of generalization curves of motor learning. Using this algorithm could lead to faster and more e cient computations and, as a result, more relevant ndings. In order to test the algorithm's performance, I run it against an algorithm which selects stimuli randomly. Eyeballing the resulting plots shows a considerable di erence in the performance, although for 2 out of 3 of the curve parameters, the results are not statistically signi cant. Future research could further build on this model to improve its performance, or use it in a model comparison study between symmetric and asymmetric models of a generalization curve.en_US
dc.thesis.facultyFaculteit der Sociale Wetenschappenen_US
dc.thesis.specialisationBachelor Artificial Intelligenceen_US
dc.thesis.studyprogrammeArtificial Intelligenceen_US
dc.titleUsing Bayesian Adaptive Stimulus Selection to Estimate Generalization Curvesen_US
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
1002844 Neascu.pdf
1.69 MB
Adobe Portable Document Format