BNEM: A Boltzmann Sampler Based on Bootstrapped Noised Energy Matching
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914378864918528 |
|---|---|
| author | OuYang, RuiKang Qiang, Bo Hernández-Lobato, José Miguel |
| author_facet | OuYang, RuiKang Qiang, Bo Hernández-Lobato, José Miguel |
| contents | Developing an efficient sampler capable of generating independent and identically distributed (IID) samples from a Boltzmann distribution is a crucial challenge in scientific research, e.g. molecular dynamics. In this work, we intend to learn neural samplers given energy functions instead of data sampled from the Boltzmann distribution. By learning the energies of the noised data, we propose a diffusion-based sampler, Noised Energy Matching, which theoretically has lower variance and more complexity compared to related works. Furthermore, a novel bootstrapping technique is applied to NEM to balance between bias and variance. We evaluate NEM and BNEM on a 2-dimensional 40 Gaussian Mixture Model (GMM) and a 4-particle double-well potential (DW-4). The experimental results demonstrate that BNEM can achieve state-of-the-art performance while being more robust. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_09787 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | BNEM: A Boltzmann Sampler Based on Bootstrapped Noised Energy Matching OuYang, RuiKang Qiang, Bo Hernández-Lobato, José Miguel Machine Learning Artificial Intelligence Computation Developing an efficient sampler capable of generating independent and identically distributed (IID) samples from a Boltzmann distribution is a crucial challenge in scientific research, e.g. molecular dynamics. In this work, we intend to learn neural samplers given energy functions instead of data sampled from the Boltzmann distribution. By learning the energies of the noised data, we propose a diffusion-based sampler, Noised Energy Matching, which theoretically has lower variance and more complexity compared to related works. Furthermore, a novel bootstrapping technique is applied to NEM to balance between bias and variance. We evaluate NEM and BNEM on a 2-dimensional 40 Gaussian Mixture Model (GMM) and a 4-particle double-well potential (DW-4). The experimental results demonstrate that BNEM can achieve state-of-the-art performance while being more robust. |
| title | BNEM: A Boltzmann Sampler Based on Bootstrapped Noised Energy Matching |
| topic | Machine Learning Artificial Intelligence Computation |
| url | https://arxiv.org/abs/2409.09787 |