Skip to Main content Skip to Navigation
New interface
Conference papers

Self-Bounding Majority Vote Learning Algorithms by the Direct Minimization of a Tight PAC-Bayesian C-Bound

Abstract : In the PAC-Bayesian literature, the C-Bound refers to an insightful relation between the risk of a majority vote classifier (under the zero-one loss) and the first two moments of its margin (i.e., the expected margin and the voters' diversity). Until now, learning algorithms developed in this framework minimize the empirical version of the C-Bound, instead of explicit PAC-Bayesian generalization bounds. In this paper, by directly optimizing PAC-Bayesian guarantees on the C-Bound, we derive self-bounding majority vote learning algorithms. Moreover, our algorithms based on gradient descent are scalable and lead to accurate predictors paired with non-vacuous guarantees.
Document type :
Conference papers
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03208948
Contributor : Paul Viallard Connect in order to contact the contributor
Submitted on : Monday, August 30, 2021 - 11:34:09 AM
Last modification on : Saturday, June 25, 2022 - 10:57:32 AM

Files

main.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03208948, version 2
  • ARXIV : 2104.13626

Collections

Citation

Paul Viallard, Pascal Germain, Amaury Habrard, Emilie Morvant. Self-Bounding Majority Vote Learning Algorithms by the Direct Minimization of a Tight PAC-Bayesian C-Bound. ECML PKDD 2021, Sep 2021, Bilbao, Spain. ⟨hal-03208948v2⟩

Share

Metrics

Record views

28

Files downloads

49