Exact performance for separable profiles
ProvedPGDRevision20261113.separable_profilesLet , and let be scalar profiles on open intervals , for . Fix one inverse stepsize , and assume that every scalar profile satisfies the hypotheses of the exact-profile theorem: twice continuous differentiability, , , , , and surjectivity of onto .
Use separable functions , , on the product interval and the actual coordinatewise implicit update. For , allow any and initial state with nonzero gradient. Define
For every finite , the exact squared Euclidean gradient ratio is
Infinite values are interpreted in the extended nonnegative reals. This is the product-profile extension of the scalar exact-performance result.
Related results. Exact profile performance (Theorem 4.1) · Geometry classification (Theorem 5.1) · Parameter localization (Theorem 7.1)
import Definitions.Def_PGDRevision20261113_Profile open PGDRevision20261113
theorem PGDRevision20261113.separable_profiles : SeparableProfilesClaim := by sorry