Inducing Permutation Invariant Priors in Bayesian Optimization for Carbon Capture and Storage Applications Journal: Journal Name

Sofianos Panagiotis Fotias, Vassilis Gaganis · 2026

Read Inducing Permutation Invariant Priors in Bayesian Optimization for Carbon Capture and Storage Applications Journal: Journal Name by Sofianos Panagiotis Fotias, Vassilis Gaganis (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-02409.txt.

Description

Sofianos Panagiotis Fotias and Vassilis Gaganis make Bayesian optimisation respect permutation symmetry in well placement for carbon capture and storage. Bayesian optimisation searches costly black-box functions with a surrogate model, usually a Gaussian process. Standard kernels suit vector inputs, not unordered sets, so they waste the symmetry when wells are run under group control and injectors or producers can be swapped.

Their kernel, GP-Perm, encodes permutation invariance by comparing sets through a stable divergence between their induced empirical representations. It can be combined with standard kernels for added vector structure. The simulator runs wells under group control, giving the symmetry the kernel exploits. The authors apply the method to well placement in storage projects.

Catalog record

CREATORSofianos Panagiotis Fotias, Vassilis Gaganis
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE81 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2605.02409

Browse the full library · About the Tower of Babel