Collection of inference methods to go from arbitrary summary statistics (neural network, peaks, power spectrum, ...) to posterior parameter constraints. Inference and neural density estimation methods ...

Abstract: We introduce a statistical inference framework for maximum likelihood estimation of the contagion source from a partially observed contagion spreading process on an arbitrary network ...

Abstract: Variational message passing is an efficient Bayesian inference method in factorized probabilistic models composed of conjugate factors from the exponential family (EF) distributions. In many ...

One of the innovations of RhythmMamba is that the model supports inference on videos of arbitrary length. I would like to ask, in practical scenarios, is it necessary for the model to support ...