Adjoint Accelerated 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
The root-finding problem is to find
Our assumptions:
Assumption 1 (Differentiability and Lipschitz
bounds). The map
Assumption 2 (Cohypomonotonicity). For every
Assumption 3 (Initial distance to a zero). There
exists
Oracle queries may use any
Proved result
OracleTemplate.Convergence LongTails.Setups.CohypomonotoneAdjoint.spec
OracleTemplate.CohypomonotoneAdjoint.AdjointAcceleratedCohypomonotone.budgetRelevant previous results
- Sucheol Lee and Donghwan Kim. Fast
Extra Gradient Methods for Smooth Structured Nonconvex-Nonconcave
Minimax Problems. 2021. Complexity:
exact operator evaluations using FEG, including initialization, for ; comparable regime . - Yang Cai, Argyris Oikonomou, and Weiqiang Zheng. Accelerated Algorithms for
Constrained Nonconvex-Nonconcave Min-Max Optimization and Comonotone
Inclusion. 2024 (ICML publication; preprint initially 2022; version
v6 used). Complexity:
exact operator evaluations using composite-EAG with the additional operator set to zero, including initialization, for and .