We introduce a new symmetric integration scheme for split Hamiltonian Monte Carlo (HMC) that enables efficient inference for Bayesian neural networks on large datasets. Our method outperforms stochastic gradient MCMC in accuracy and uncertainty quantification, demonstrating HMC as a viable option for large-scale machine learning problems.
Dec 1, 2021
Dec 1, 2021
Jul 1, 2021
Jun 1, 2021
Jan 1, 2021
Jan 1, 2021
Jan 1, 2021
Jan 1, 2020
Jan 1, 2020
Dec 1, 2019