Hamiltonian Monte Carlo
A Markov chain Monte Carlo sampling algorithm that uses Hamiltonian dynamics simulation to efficiently explore a target distribution. It is widely applied in Bayesian inference and statistical physics.
Timeline (last 90 days)
The classical Dobrushin contraction version requires globally uniform contraction, which is too stringent for algorithms like HMC that mix quickly only locally and may be very slow globally.
Hamiltonian Monte Carlo (HMC) is widely used in Bayesian inference and statistical physics because of its dual advantages in convergence speed with respect to condition number and dimension.
Rigorously proving HMC's acceleration in terms of both condition number and dimension simultaneously had long been an open problem.
Previous studies (such as those by Bou-Rabee and Chen et al.) demonstrated acceleration in either condition number or dimension separately, but could not prove both rigorously at the same time.
MCMC and Hamiltonian Monte Carlo (HMC) are modern variants of the Monte Carlo method that improve sampling efficiency and enable Bayesian methods to be applied to high-dimensional complex models.
哈密顿蒙特卡洛(HMC)方法将经典力学中的哈密顿力学引入概率采样,在贝叶斯深度学习和概率编程语言(如Stan、PyMC)中被广泛使用
哈密顿蒙特卡洛(HMC)方法在贝叶斯深度学习和概率编程语言(如Stan、PyMC)中被广泛使用
模拟退火算法借鉴了冶金学中的退火工艺,哈密顿蒙特卡洛方法将经典力学中的哈密顿力学引入概率采样
All Facts (8)
The classical Dobrushin contraction version requires globally uniform contraction, which is too stringent for algorithms like HMC that mix quickly only locally and may be very slow globally.
50%UnverifiedHamiltonian Monte Carlo (HMC) is widely used in Bayesian inference and statistical physics because of its dual advantages in convergence speed with respect to condition number and dimension.
50%UnverifiedRigorously proving HMC's acceleration in terms of both condition number and dimension simultaneously had long been an open problem.
50%UnverifiedPrevious studies (such as those by Bou-Rabee and Chen et al.) demonstrated acceleration in either condition number or dimension separately, but could not prove both rigorously at the same time.
50%UnverifiedMCMC and Hamiltonian Monte Carlo (HMC) are modern variants of the Monte Carlo method that improve sampling efficiency and enable Bayesian methods to be applied to high-dimensional complex models.
50%Unverified哈密顿蒙特卡洛(HMC)方法将经典力学中的哈密顿力学引入概率采样,在贝叶斯深度学习和概率编程语言(如Stan、PyMC)中被广泛使用
50%Unverified哈密顿蒙特卡洛(HMC)方法在贝叶斯深度学习和概率编程语言(如Stan、PyMC)中被广泛使用
50%Unverified模拟退火算法借鉴了冶金学中的退火工艺,哈密顿蒙特卡洛方法将经典力学中的哈密顿力学引入概率采样
50%