Mitigating mode collapse in normalizing flows by annealing with an adaptive schedule: Application to parameter estimation
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918011534835712 |
|---|---|
| author | Wang, Yihang Chi, Chris Dinner, Aaron R. |
| author_facet | Wang, Yihang Chi, Chris Dinner, Aaron R. |
| contents | Normalizing flows (NFs) provide uncorrelated samples from complex distributions, making them an appealing tool for parameter estimation. However, the practical utility of NFs remains limited by their tendency to collapse to a single mode of a multimodal distribution. In this study, we show that annealing with an adaptive schedule based on the effective sample size (ESS) can mitigate mode collapse. We demonstrate that our approach can converge the marginal likelihood for a biochemical oscillator model fit to time-series data in ten-fold less computation time than a widely used ensemble Markov chain Monte Carlo (MCMC) method. We show that the ESS can also be used to reduce variance by pruning the samples. We expect these developments to be of general use for sampling with NFs and discuss potential opportunities for further improvements. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_03652 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Mitigating mode collapse in normalizing flows by annealing with an adaptive schedule: Application to parameter estimation Wang, Yihang Chi, Chris Dinner, Aaron R. Machine Learning Computational Physics Data Analysis, Statistics and Probability Quantitative Methods Normalizing flows (NFs) provide uncorrelated samples from complex distributions, making them an appealing tool for parameter estimation. However, the practical utility of NFs remains limited by their tendency to collapse to a single mode of a multimodal distribution. In this study, we show that annealing with an adaptive schedule based on the effective sample size (ESS) can mitigate mode collapse. We demonstrate that our approach can converge the marginal likelihood for a biochemical oscillator model fit to time-series data in ten-fold less computation time than a widely used ensemble Markov chain Monte Carlo (MCMC) method. We show that the ESS can also be used to reduce variance by pruning the samples. We expect these developments to be of general use for sampling with NFs and discuss potential opportunities for further improvements. |
| title | Mitigating mode collapse in normalizing flows by annealing with an adaptive schedule: Application to parameter estimation |
| topic | Machine Learning Computational Physics Data Analysis, Statistics and Probability Quantitative Methods |
| url | https://arxiv.org/abs/2505.03652 |