A Framework for Bounding Deterministic Risk with PAC-Bayes: Applications to Majority Votes
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915585478098944 |
|---|---|
| author | Leblanc, Benjamin Germain, Pascal |
| author_facet | Leblanc, Benjamin Germain, Pascal |
| contents | PAC-Bayes is a popular and efficient framework for obtaining generalization guarantees in situations involving uncountable hypothesis spaces. Unfortunately, in its classical formulation, it only provides guarantees on the expected risk of a randomly sampled hypothesis. This requires stochastic predictions at test time, making PAC-Bayes unusable in many practical situations where a single deterministic hypothesis must be deployed. We propose a unified framework to extract guarantees holding for a single hypothesis from stochastic PAC-Bayesian guarantees. We present a general oracle bound and derive from it a numerical bound and a specialization to majority vote. We empirically show that our approach consistently outperforms popular baselines (by up to a factor of 2) when it comes to generalization bounds on deterministic classifiers. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_25569 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Framework for Bounding Deterministic Risk with PAC-Bayes: Applications to Majority Votes Leblanc, Benjamin Germain, Pascal Machine Learning PAC-Bayes is a popular and efficient framework for obtaining generalization guarantees in situations involving uncountable hypothesis spaces. Unfortunately, in its classical formulation, it only provides guarantees on the expected risk of a randomly sampled hypothesis. This requires stochastic predictions at test time, making PAC-Bayes unusable in many practical situations where a single deterministic hypothesis must be deployed. We propose a unified framework to extract guarantees holding for a single hypothesis from stochastic PAC-Bayesian guarantees. We present a general oracle bound and derive from it a numerical bound and a specialization to majority vote. We empirically show that our approach consistently outperforms popular baselines (by up to a factor of 2) when it comes to generalization bounds on deterministic classifiers. |
| title | A Framework for Bounding Deterministic Risk with PAC-Bayes: Applications to Majority Votes |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2510.25569 |