DeepAI AI Chat
Log In Sign Up

Model Selection for Simulator-based Statistical Models: A Kernel Approach

by   Takafumi Kajihara, et al.
Max Planck Society
nec global

We propose a novel approach to model selection for simulator-based statistical models. The proposed approach defines a mixture of candidate models, and then iteratively updates the weight coefficients for those models as well as the parameters in each model simultaneously; this is done by recursively applying Bayes' rule, using the recently proposed kernel recursive ABC algorithm. The practical advantage of the method is that it can be used even when a modeler lacks appropriate prior knowledge about the parameters in each model. We demonstrate the effectiveness of the proposed approach with a number of experiments, including model selection for dynamical systems in ecology and epidemiology.


A Nonconformity Approach to Model Selection for SVMs

We investigate the issue of model selection and the use of the nonconfor...

Online simulator-based experimental design for cognitive model selection

The problem of model selection with a limited number of experimental tri...

A Local Information Criterion for Dynamical Systems

Encoding a sequence of observations is an essential task with many appli...

Kernel Recursive ABC: Point Estimation with Intractable Likelihood

We propose a novel approach to parameter estimation for simulator-based ...

Evaluation of Model Selection for Kernel Fragment Recognition in Corn Silage

Model selection when designing deep learning systems for specific use-ca...

Closed-loop Model Selection for Kernel-based Models using Bayesian Optimization

Kernel-based nonparametric models have become very attractive for model-...