Descriptions may be stale. Consult the linked Lean sources.
Finite-Sum
Weak Minty Inclusions
We consider finding with
in a real Hilbert
space , where averages component maps and is monotone. The
average is Lipschitz, component differences satisfy a mean-square
Lipschitz bound, and a weak Minty solution lies within a prescribed
distance of the given starting point. The oracle returns exact component
values and resolvent evaluations of . For , the goal is a
measurable pair with almost surely and ,
where is the law of the
algorithm’s randomness.
More precisely, the inclusion is Write for the inner
product on and for its induced norm. The
parameters satisfy The algorithm
receives and the
starting point ; the maps
and are accessed through the oracle.
Our assumptions:
Assumption 1 (Monotonicity and resolvent access).
For every , , and , There is a
map ,
written , such that
Assumption 2 (Finite-sum Lipschitz bounds). For all
,
Assumption 3 (Weak Minty solution and initial
distance). There exists such that and, for every and ,
A component query returns
for any and . A resolvent query returns for any and ; the stated resolvent condition
applies when . Each query
of either kind counts as one call toward a common oracle-call
budget.
After the allotted calls, the output must be measurable, its
squared residual must be integrable under , and it must satisfy