AI RESEARCH

Hyperspherical Variational Autoencoders Using Efficient Spherical Cauchy Distribution

arXiv CS.AI

ArXi:2506.21278v3 Announce Type: replace-cross We propose spherical Cauchy (spCauchy) latent variables for variational autoencoders on hyperspherical latent spaces. The spCauchy family has heavy-tailed global behavior and admits an exact differentiable reparameterization by applying a M\"obius transformation to uniform samples on the sphere.