Adding solver sensitivity information to the training loss of optimization proxies reduces prediction error and constraint violations on AC-OPF benchmarks and improves self-supervised portfolio proxies in the medium-risk regime, though the theory and experiments only partially align.
Sensitivity analysis for nonlinear programming using penalty methods
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Sobolev Training of End-to-End Optimization Proxies
Adding solver sensitivity information to the training loss of optimization proxies reduces prediction error and constraint violations on AC-OPF benchmarks and improves self-supervised portfolio proxies in the medium-risk regime, though the theory and experiments only partially align.