At the symmetrization interface in VC learning, the ghost-gap bad event is analytic (hence null-measurable w.r.t. finite Borel measures) for Borel-parameterized classes on Polish spaces, with a strict separation from Borel and closure under natural constructors.
Formal Learning Theory Kernel: a Lean 4 formalization of the fundamental theorem of statistical learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Null Measurability at the Symmetrization Interface in VC Learning
At the symmetrization interface in VC learning, the ghost-gap bad event is analytic (hence null-measurable w.r.t. finite Borel measures) for Borel-parameterized classes on Polish spaces, with a strict separation from Borel and closure under natural constructors.