Under squared loss, the relative regret of an unbiased learner is an upper bound on the squared correlation between its actual error and any unbiased error assessor.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.ST 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A Heisenberg-esque Uncertainty Principle for Simultaneous (Machine) Learning and Error Assessment?
Under squared loss, the relative regret of an unbiased learner is an upper bound on the squared correlation between its actual error and any unbiased error assessor.