Paper gives exact global minimax risk near nonidentification
Conditions let one estimator match the leading constant uniformly; two models fix fifth- and sixth-order risk constants.
2 stories, newest first · one card per story
Conditions let one estimator match the leading constant uniformly; two models fix fifth- and sixth-order risk constants.
The framework covers score matching, ratio matching, pseudolikelihood and noise contrastive estimation, then characterizes estimator efficiency.
Today 1.6K readers · 2.4K views · 774 reads · all time 189K views