When does Metropolized Hamiltonian Monte Carlo provably outperform Metropolis-adjusted Langevin algorithm?
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908827758100480 |
|---|---|
| author | Chen, Yuansi Gatmiry, Khashayar Jiang, Minhui |
| author_facet | Chen, Yuansi Gatmiry, Khashayar Jiang, Minhui |
| contents | We analyze the mixing time of Metropolized Hamiltonian Monte Carlo (HMC) with the leapfrog integrator to sample from a distribution on $\mathbb{R}^d$ whose log-density is smooth, has Lipschitz Hessian in Frobenius norm and satisfies isoperimetry. We bound the gradient complexity to reach $ε$ error in total variation distance from a warm start by $\tilde O(d^{1/4}\text{polylog}(1/ε))$ and demonstrate the benefit of choosing the number of leapfrog steps to be larger than 1. To surpass the previous analysis on Metropolis-adjusted Langevin algorithm (MALA) that has $\tilde{O}(d^{1/2}\text{polylog}(1/ε))$ dimension dependency [WSC22], we reveal a key feature in our proof that the joint distribution of the location and velocity variables of the discretization of the continuous HMC dynamics stays approximately invariant. This key feature, when shown via induction over the number of leapfrog steps, enables us to obtain estimates on moments of various quantities that appear in the acceptance rate control of Metropolized HMC. Notably, our analysis does not require log-concavity or independence of the marginals, and only relies on an isoperimetric inequality. To illustrate the relevance of the Lipschitz Hessian in Frobenius norm assumption, several examples that fall into our framework are discussed. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2304_04724 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | When does Metropolized Hamiltonian Monte Carlo provably outperform Metropolis-adjusted Langevin algorithm? Chen, Yuansi Gatmiry, Khashayar Jiang, Minhui Computation Computational Complexity Machine Learning We analyze the mixing time of Metropolized Hamiltonian Monte Carlo (HMC) with the leapfrog integrator to sample from a distribution on $\mathbb{R}^d$ whose log-density is smooth, has Lipschitz Hessian in Frobenius norm and satisfies isoperimetry. We bound the gradient complexity to reach $ε$ error in total variation distance from a warm start by $\tilde O(d^{1/4}\text{polylog}(1/ε))$ and demonstrate the benefit of choosing the number of leapfrog steps to be larger than 1. To surpass the previous analysis on Metropolis-adjusted Langevin algorithm (MALA) that has $\tilde{O}(d^{1/2}\text{polylog}(1/ε))$ dimension dependency [WSC22], we reveal a key feature in our proof that the joint distribution of the location and velocity variables of the discretization of the continuous HMC dynamics stays approximately invariant. This key feature, when shown via induction over the number of leapfrog steps, enables us to obtain estimates on moments of various quantities that appear in the acceptance rate control of Metropolized HMC. Notably, our analysis does not require log-concavity or independence of the marginals, and only relies on an isoperimetric inequality. To illustrate the relevance of the Lipschitz Hessian in Frobenius norm assumption, several examples that fall into our framework are discussed. |
| title | When does Metropolized Hamiltonian Monte Carlo provably outperform Metropolis-adjusted Langevin algorithm? |
| topic | Computation Computational Complexity Machine Learning |
| url | https://arxiv.org/abs/2304.04724 |