Simulation of the von mises fisher distribution

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Generating Sentences by Editing Prototypes (2017, methodology): “This distribution is straightforward to use as part of a variational autoen-coder, as sampling can be easily performed using the reparameterization trick and the rejection sampler of Wood (1994).”
A Bayesian Analysis of Directional Data Using the von Mises–Fisher Distribution (2005, methodology): “In higher dimensions p > 3 , one can use Wood’s method (Wood, 1994) which is based on the VM algorithm of Ulrich (1984).”
Clustering on the Unit Hypersphere using von Mises-Fisher Distributions (2005, methodology): “We used a slight modification of the algorithm given by Wood (1994) to generate a set of data points followinga given vMF distribution.”
Spherical Latent Spaces for Stable Variational Autoencoders (2018, methodology): “Sampling from vMF Following the implementation of Guu et al. (2018), we use the rejection sampling scheme of Wood (1994) to sample a “change magnitude” w .”
Learning grasp affordance densities (2011, methodology): “E cient simulation methods are available for both normal distributions [2] and vonMises Fisher distributions [34].”
SIMULATING BESSEL RANDOM VARIABLES (2002, methodology): “A uniformly fast algorithm for this distribution was derived by Best and Fisher (1979), with alternate methods proposed later by Dagpunar (1990), Barabesi (1993) and Wood (1994).”