ADELIA is the first AD-enabled INLA system that computes exact hyperparameter gradients via a structure-exploiting multi-GPU backward pass, delivering 4.2-7.9x per-gradient speedups and 5-8x better energy efficiency than finite differences on models with up to 1.9 million latent variables.
mpi4jax: Zero-copy MPI communication of JAX arrays
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
method 2
citation-polarity summary
years
2026 2roles
method 2polarities
use method 2representative citing papers
Neural quantum states on K5 yield two families of approximate physical states for the Thiemann-ordered Hamiltonian constraint in Abelianized Euclidean LQG: one flat with non-zero volume (non-normalizable) and one normalizable with zero volume, close to Ashtekar-Lewandowski and Dittrich-Geiller vacua
citing papers explorer
-
ADELIA: Automatic Differentiation for Efficient Laplace Inference Approximations
ADELIA is the first AD-enabled INLA system that computes exact hyperparameter gradients via a structure-exploiting multi-GPU backward pass, delivering 4.2-7.9x per-gradient speedups and 5-8x better energy efficiency than finite differences on models with up to 1.9 million latent variables.
-
Finding and characterising physical states of Euclidean Abelianized loop quantum gravity using neural quantum states
Neural quantum states on K5 yield two families of approximate physical states for the Thiemann-ordered Hamiltonian constraint in Abelianized Euclidean LQG: one flat with non-zero volume (non-normalizable) and one normalizable with zero volume, close to Ashtekar-Lewandowski and Dittrich-Geiller vacua