Flow-Based Density Ratio Estimation for Intractable Distributions with Applications in Genomics
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911735500242944 |
|---|---|
| author | Antipov, Egor Palma, Alessandro Consoli, Lorenzo Günnemann, Stephan Dittadi, Andrea Theis, Fabian J. |
| author_facet | Antipov, Egor Palma, Alessandro Consoli, Lorenzo Günnemann, Stephan Dittadi, Andrea Theis, Fabian J. |
| contents | Estimating density ratios between pairs of intractable data distributions is a core problem in probabilistic modeling, enabling principled comparisons of sample likelihoods under different data-generating processes across conditions. While exact-likelihood models such as normalizing flows offer a promising approach to density ratio estimation, naive evaluations are computationally expensive and prone to discretization errors because they require simulating each distribution's likelihood independently. In this work, we leverage condition-aware flow matching to derive a single dynamical formulation for tracking density ratios along generative trajectories. We demonstrate competitive performance on simulated benchmarks for closed-form ratio estimation, and show that our method supports versatile tasks in single-cell genomics data analysis, where likelihood-based comparisons of cellular states across experimental conditions enable treatment effect estimation and batch correction evaluation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_24201 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Flow-Based Density Ratio Estimation for Intractable Distributions with Applications in Genomics Antipov, Egor Palma, Alessandro Consoli, Lorenzo Günnemann, Stephan Dittadi, Andrea Theis, Fabian J. Machine Learning Estimating density ratios between pairs of intractable data distributions is a core problem in probabilistic modeling, enabling principled comparisons of sample likelihoods under different data-generating processes across conditions. While exact-likelihood models such as normalizing flows offer a promising approach to density ratio estimation, naive evaluations are computationally expensive and prone to discretization errors because they require simulating each distribution's likelihood independently. In this work, we leverage condition-aware flow matching to derive a single dynamical formulation for tracking density ratios along generative trajectories. We demonstrate competitive performance on simulated benchmarks for closed-form ratio estimation, and show that our method supports versatile tasks in single-cell genomics data analysis, where likelihood-based comparisons of cellular states across experimental conditions enable treatment effect estimation and batch correction evaluation. |
| title | Flow-Based Density Ratio Estimation for Intractable Distributions with Applications in Genomics |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2602.24201 |