Introduces α-LIS, a provable generalization of likelihood-informed subspaces to α-tempered posteriors with practical extensions for limited noisy data and unavailable gradients.
Hyperparameter optimization for randomized algorithms: a case study on random features
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.CO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Likelihood-informed dimension reduction across tempered Bayesian posteriors
Introduces α-LIS, a provable generalization of likelihood-informed subspaces to α-tempered posteriors with practical extensions for limited noisy data and unavailable gradients.