Averaging a model over a symmetry group with a data-dependent kernel shrinks the KL term in PAC-Bayes bounds, giving tighter guarantees for non-compact groups and non-invariant data.
User-Friendly Introduction to PAC-Bayes Bounds
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Symmetries in PAC-Bayesian Learning
Averaging a model over a symmetry group with a data-dependent kernel shrinks the KL term in PAC-Bayes bounds, giving tighter guarantees for non-compact groups and non-invariant data.