Optimizes a Neural Radiance Field via probability density distillation from a 2D diffusion model to produce text-conditioned 3D scenes viewable from any angle.
Sticking the landing: Simple, lower-variance gradient estimators for variational inference
2 Pith papers cite this work, alongside 73 external citations. Polarity classification is still indexing.
2
Pith papers citing it
73
external citations · Pith
abstract
We propose a simple and general variant of the standard reparameterized gradient estimator for the variational evidence lower bound. Specifically, we remove a part of the total derivative with respect to the variational parameters that corresponds to the score function. Removing this term produces an unbiased gradient estimator whose variance approaches zero as the approximate posterior approaches the exact posterior. We analyze the behavior of this gradient estimator theoretically and empirically, and generalize it to more complex variational distributions such as mixtures and importance-weighted posteriors.
representative citing papers
A finite-element variational inference method delivers full-covariance Bayesian field reconstruction at dimensions exceeding 400,000 for 3D porous media flow using sparse precision parameterization from SPDE priors.
citing papers explorer
-
DreamFusion: Text-to-3D using 2D Diffusion
Optimizes a Neural Radiance Field via probability density distillation from a 2D diffusion model to produce text-conditioned 3D scenes viewable from any angle.
-
Scalable High-Dimensional Bayesian Field Reconstruction with Finite Elements: Application to 3D Porous Media Flow
A finite-element variational inference method delivers full-covariance Bayesian field reconstruction at dimensions exceeding 400,000 for 3D porous media flow using sparse precision parameterization from SPDE priors.