Self-Supervised Tuning of Low-Fidelity Optimization Models via High-Fidelity Feedback
Abstract
Low-fidelity (LF) optimization models are widely used in large-scale operations because they can be solved efficiently and reliably, but they often exhibit systematic errors relative to high-fidelity (HF) formulations that better capture physical and operational constraints. Existing learning-based approaches typically rely on supervised or semi-supervised training with HF solutions, which are expensive to compute and may not align with HF operational objectives. We propose a self-supervised framework that parameterizes LF optimization models directly from HF feedback. The low-fidelity problem is embedded as a differentiable optimization layer and trained end-to-end, while a HF correction problem evaluates its solution and provides gradient information through its dual variables, without requiring HF solution labels. We evaluate the framework on optimal power flow, optimal gas flow, and pooling problems, spanning different sources of low- and high-fidelity mismatch. Across all three applications, the self-supervised approach improves objective quality and feasibility over untuned LF models and generally outperforms semi-supervised training. The learned parameterizations also generalize to unseen operating conditions and, for network applications with up to 30,000 nodes, to systems not encountered during training.
Related
- HyperMC: Multi-Fidelity Hyperparameter Tuning for Stochastic Gradient MCMC
- Feedback-Aware Tuning of Recursive Q-Learning
- MF-SCBO : Multi-fidelity Scalable Constrained Bayesian Optimization
- Reduced-Space Multi-Fidelity Bayesian Optimization of Process Simulation Models
- Cost-Aware Iteration Allocation for Sequential Multi-Fidelity First-Order Optimization
BibTeX
@misc{constanteflores2026selfsupervised,
title = {{Self-Supervised Tuning of Low-Fidelity Optimization Models via High-Fidelity Feedback}},
author = {Gonzalo E. Constante-Flores and André H. Quisaguano-Paredes and Asha Ramanujam and Rishabh Gupta and Can Li},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.28851},
eprint = {2610.28851},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.28851},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Self-Supervised Tuning of Low-Fidelity Optimization Models via High-Fidelity Feedback
AU - Gonzalo E. Constante-Flores
AU - André H. Quisaguano-Paredes
AU - Asha Ramanujam
AU - Rishabh Gupta
AU - Can Li
PY - 2026
DA - 2026/10/20
PB - arXiv
DO - 10.48550/arxiv.2610.28851
UR - https://arxiv.org/abs/2610.28851
AB - Low-fidelity (LF) optimization models are widely used in large-scale operations because they can be solved efficiently and reliably, but they often exhibit systematic errors relative to high-fidelity (HF) formulations that better capture physical and operational constraints. Existing learning-based approaches typically rely on supervised or semi-supervised training with HF solutions, which are expensive to compute and may not align with HF operational objectives. We propose a self-supervised framework that parameterizes LF optimization models directly from HF feedback. The low-fidelity problem is embedded as a differentiable optimization layer and trained end-to-end, while a HF correction problem evaluates its solution and provides gradient information through its dual variables, without requiring HF solution labels. We evaluate the framework on optimal power flow, optimal gas flow, and pooling problems, spanning different sources of low- and high-fidelity mismatch. Across all three applications, the self-supervised approach improves objective quality and feasibility over untuned LF models and generally outperforms semi-supervised training. The learned parameterizations also generalize to unseen operating conditions and, for network applications with up to 30,000 nodes, to systems not encountered during training.
ER -