Establishes measure-theoretic foundations for NML in regular non-smooth models and introduces the PDL-PPMH geometric MCMC sampler to compute stochastic complexity exactly.
To ensure the chain does not get “stuck” (which would destroy the spectral gap), the variance of the randomized acceptance ratio must be strictly bounded
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
The Normalized Maximum Likelihood for Regular Non-Smooth Models: Measure-Theoretic Foundations and Geometric Sampling
Establishes measure-theoretic foundations for NML in regular non-smooth models and introduces the PDL-PPMH geometric MCMC sampler to compute stochastic complexity exactly.