Reconstructing the universe with variational self-boosted sampling

Reconstructing the Universe

We provide detailed evidence for the claim that nonperturbative quantum gravity, defined through state sums of causal triangulated geometries, possesses a large-scale limit in which the dimension of spacetime is four and the dynamics of the volume of the universe behaves semiclassically. This is a first step in reconstructing the universe from a dynamical principle at the Planck scale, and at t...

On sampling and reconstructing surfaces with boundaries

Surface reconstruction algorithms can ensure correctness of the reconstructed surface by imposing conditions on sampling and by assuming that the given point set satisfies the prescribed conditions. Traditionally, such sampling conditions impose minimum required sampling density for correct reconstruction. Let us assume that no additional information is provided to the reconstruction algorithm ...

Sampling and Reconstructing User Experience

The Experience Sampling and Reconstruction Method (ESRM) is a research method suitable for user studies conducted in situ that is needed for the design and evaluation of ambient intelligence technologies. ESRM is a diary method supported by a distributed application, Reconexp, which runs on a mobile device and a website, enabling surveying user attitudes, experiences, and requirements in field ...

The Self-organizing Universe

Variational Rejection Sampling

Learning latent variable models with stochastic variational inference is challenging when the approximate posterior is far from the true posterior, due to high variance in the gradient estimates. We propose a novel rejection sampling step that discards samples from the variational posterior which are assigned low likelihoods by the model. Our approach provides an arbitrarily accurate approximat...