Constrained Generalized Monotone — Constrained Generalized Monotone Inequalities
Descriptions may be stale. Consult the linked Lean sources.
Constrained Generalized Monotone Inequalities · Lean proof
Constrained
Generalized Monotone Inequalities
We consider a variational inequality for a continuous monotone
operator on a convex set
in a real inner product space
. The operator satisfies a
segmentwise generalized Lipschitz condition with constants and . A given point has bounded distance to a
solution, with operator norms bounded at both points. Two exact oracles
return and the projection of
onto . For , the goal is a feasible
output admitting with for
every and , almost
surely under , the law of the
algorithm’s randomness.
The variational inequality asks for such that The norm is induced by the inner product
on . The parameters satisfy The algorithm
receives only , , , and ; the parameters , , , and may enter the oracle-call
budgets but are not algorithm inputs.
Our assumptions:
Assumption 1 (Convex constraints and projection).
The set is convex.
There exists a map
satisfying, for every ,
Assumption 2 (Monotonicity and generalized Lipschitz
continuity). The operator is
continuous on and satisfies For every and , the segment bound implies
Assumption 3 (Initial point and solution bounds).
The given starting point satisfies and .
There exists a solution of
the variational inequality such that
Queries to either oracle may be made at any point of . The operator oracle returns at , and the projection oracle returns
at . Calls to these two oracles are counted
separately.
For a feasible point , define
its normal cone by After the allotted oracle calls,
the output must satisfy the normal-cone residual criterion
Proved result
OracleTemplate.Convergence LongTails.Setups.ConstrainedGeneralizedMonotone.spec
OracleTemplate.ConstrainedMonotone.budget
Relevant previous results
No qualifying prior reference verified; this is not a proof of novelty.