A tutorial on particle filters

This tutorial aims to provide an accessible introduction to particle filters, and sequential Monte Carlo (SMC) more generally. These techniques allow for Bayesian inference in complex dynamic state-space models and have become increasingly popular over the last decades. The basic building blocks of SMC-sequential importance sampling and resampling-are discussed in detail with illustrative examples. A final example presents a particle filter for estimating time-varying learning rates in a probabilistic category learning task.