Adjoint Anchored Cohypomonotone — Cohypomonotone Root Finding with Adjoints
Descriptions may be stale. Consult the linked Lean sources.
Cohypomonotone Root Finding with Adjoints · Lean proof
Cohypomonotone Root
Finding with Adjoints
We seek a zero of a differentiable map on a finite-dimensional real
inner product space . The map is
-Lipschitz and -cohypomonotone, and its derivative
is -Lipschitz, where , , and . A given point lies within distance of a zero. The oracle returns
or the adjoint product for . For , the goal is an output
satisfying almost
surely under , the law of the
algorithm’s randomness.
The root-finding problem is to find such that The
norm is induced by the
inner product on , and denotes the
induced operator norm. The algorithm receives only , , , , and ; the radius and tolerance may enter the oracle-call
budget but are not algorithm inputs.
Our assumptions:
Assumption 1 (Differentiability and Lipschitz
bounds). The map is Fréchet
differentiable at every ,
with derivative . For every
,
Assumption 2 (Cohypomonotonicity). For every ,
Assumption 3 (Initial distance to a zero). There
exists such that
Oracle queries may use any . Both reply types are exact, with the adjoint characterized
by Each evaluation of or counts as one call toward a
shared oracle-call budget. After the allotted calls, the output must
satisfy
Proved result
OracleTemplate.Convergence LongTails.Setups.CohypomonotoneAdjoint.spec
OracleTemplate.CohypomonotoneAdjoint.AdjointAnchoredCohypomonotone.budget
Relevant previous results
No qualifying prior reference verified; this is not a proof of novelty.