Categorical estimation task: Transient performance after changes. (A) At each time step, the agent sees one out of five possible categories (black dots) drawn from a categorical distribution with parameters . Occasional abrupt changes happen with probability and are marked with red lines. After each change, a new vector is drawn from a Dirichlet distribution with stochasticity parameter . In this example, and . (B) Mean squared error for the estimation of at each time step after an environmental change, that is, the average of over time; (left panel) and (right panel). The shaded area corresponds to the standard error of the mean. Abbreviations: pf: Particle Filtering with particles; MP: Message Passing with particles; VarSMiLe: Variational SMiLe; SOR: Stratisfied Optimal Resampling with particles (Fearnhead & Liu, 2007); SMiLe: Faraji et al. (2018); Leaky: Leaky Integrator; Exact Bayes: Adams and MacKay (2007).
This site uses cookies. By continuing to use our website, you are agreeing to our privacy policy. No content on this site may be used to train artificial intelligence systems without permission in writing from the MIT Press.