A hybrid VI-HMC method runs expensive Hamiltonian Monte Carlo only on the neural network parameters most responsible for predictive uncertainty, cutting sampling cost while approximating full HMC posteriors.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Accelerating Hamiltonian Monte Carlo for Bayesian Inference in Neural Networks and Neural Operators
A hybrid VI-HMC method runs expensive Hamiltonian Monte Carlo only on the neural network parameters most responsible for predictive uncertainty, cutting sampling cost while approximating full HMC posteriors.