Constraint-Guided Prediction Refinement via Deterministic Diffusion Trajectories

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Dogoulis, Pantelis, Bernier, Fabien, Fourreau, Félix, Tit, Karim, Cordy, Maxime
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914164654473216
author Dogoulis, Pantelis
Bernier, Fabien
Fourreau, Félix
Tit, Karim
Cordy, Maxime
author_facet Dogoulis, Pantelis
Bernier, Fabien
Fourreau, Félix
Tit, Karim
Cordy, Maxime
contents Many real-world machine learning tasks require outputs that satisfy hard constraints, such as physical conservation laws, structured dependencies in graphs, or column-level relationships in tabular data. Existing approaches rely either on domain-specific architectures and losses or on strong assumptions on the constraint space, restricting their applicability to linear or convex constraints. We propose a general-purpose framework for constraint-aware refinement that leverages denoising diffusion implicit models (DDIMs). Starting from a coarse prediction, our method iteratively refines it through a deterministic diffusion trajectory guided by a learned prior and augmented by constraint gradient corrections. The approach accommodates a wide class of non-convex and nonlinear equality constraints and can be applied post hoc to any base model. We demonstrate the method in two representative domains: constrained adversarial attack generation on tabular data with column-level dependencies and in AC power flow prediction under Kirchhoff's laws. Across both settings, our diffusion-guided refinement improves both constraint satisfaction and performance while remaining lightweight and model-agnostic.
format Preprint
id arxiv_https___arxiv_org_abs_2506_12911
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Constraint-Guided Prediction Refinement via Deterministic Diffusion Trajectories
Dogoulis, Pantelis
Bernier, Fabien
Fourreau, Félix
Tit, Karim
Cordy, Maxime
Artificial Intelligence
Machine Learning
Many real-world machine learning tasks require outputs that satisfy hard constraints, such as physical conservation laws, structured dependencies in graphs, or column-level relationships in tabular data. Existing approaches rely either on domain-specific architectures and losses or on strong assumptions on the constraint space, restricting their applicability to linear or convex constraints. We propose a general-purpose framework for constraint-aware refinement that leverages denoising diffusion implicit models (DDIMs). Starting from a coarse prediction, our method iteratively refines it through a deterministic diffusion trajectory guided by a learned prior and augmented by constraint gradient corrections. The approach accommodates a wide class of non-convex and nonlinear equality constraints and can be applied post hoc to any base model. We demonstrate the method in two representative domains: constrained adversarial attack generation on tabular data with column-level dependencies and in AC power flow prediction under Kirchhoff's laws. Across both settings, our diffusion-guided refinement improves both constraint satisfaction and performance while remaining lightweight and model-agnostic.
title Constraint-Guided Prediction Refinement via Deterministic Diffusion Trajectories
topic Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2506.12911