MIRA: A Score for Conditional Distribution Accuracy and Model Comparison
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914527454429184 |
|---|---|
| author | Sharief, Sammy Zeghal, Justine Barco, Gabriel Missael Lemos, Pablo Hezaveh, Yashar Perreault-Levasseur, Laurence |
| author_facet | Sharief, Sammy Zeghal, Justine Barco, Gabriel Missael Lemos, Pablo Hezaveh, Yashar Perreault-Levasseur, Laurence |
| contents | We introduce Mira, a sample-based score for assessing the accuracy of a candidate conditional distribution using only joint samples from the true data-generating process. Relying on the principle that distributions coincide if they assign equal probability mass to all regions, we derive an analytic expression for the Mira statistic, whose average defines the Mira score. This formulation further allows us to compute theoretical reference values and uncertainty estimates when the candidate distribution matches the true one. This framework enables model comparison by quantifying the alignment between the conditional distribution of a candidate model and the true data generating process. Consequently, Mira enables Bayesian model comparison through direct posterior validation, bypassing the challenging evidence computation. We demonstrate its effectiveness across several toy problems and Bayesian inference tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_02014 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MIRA: A Score for Conditional Distribution Accuracy and Model Comparison Sharief, Sammy Zeghal, Justine Barco, Gabriel Missael Lemos, Pablo Hezaveh, Yashar Perreault-Levasseur, Laurence Machine Learning We introduce Mira, a sample-based score for assessing the accuracy of a candidate conditional distribution using only joint samples from the true data-generating process. Relying on the principle that distributions coincide if they assign equal probability mass to all regions, we derive an analytic expression for the Mira statistic, whose average defines the Mira score. This formulation further allows us to compute theoretical reference values and uncertainty estimates when the candidate distribution matches the true one. This framework enables model comparison by quantifying the alignment between the conditional distribution of a candidate model and the true data generating process. Consequently, Mira enables Bayesian model comparison through direct posterior validation, bypassing the challenging evidence computation. We demonstrate its effectiveness across several toy problems and Bayesian inference tasks. |
| title | MIRA: A Score for Conditional Distribution Accuracy and Model Comparison |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2605.02014 |