Interpretable Neural Marked Statistics for Cosmological Inference

June 09, 2026 Β· Grace Period Β· πŸ› ICML 2026

⏳ Grace Period
This paper is less than 90 days old. We give authors time to release their code before passing judgment.
Authors Federico Semenzato, Benjamin D. Wandelt, Michele Liguori, Alvise Raccanelli arXiv ID 2606.11295 Category astro-ph.CO Cross-listed cs.LG Citations 0 Venue ICML 2026
Abstract
Recovering cosmological information beyond the power spectrum is a central goal for upcoming cosmological surveys, since late-time non-Gaussian signal in the matter density cannot be accessed through two-point statistics alone. Marked statistics fold part of this information back into the two-point level by reweighting the field with non-linear functions. We propose a neural marking scheme to generalize this process through a set of interpretable, physically motivated transformations that directly allow to interpret the gain in cosmological information at the morphological level. We employ a contrastive learning objective to align learnable marked summaries with the underlying cosmological parameters. At $k_{\max}=0.2\,h\mathrm{Mpc}^{-1}$, our neural mark tightens the marginalized constraint on $Οƒ_8$ by $2.9\times$ and on $Ξ©_m$ by $1.8\times$ compared to classical marks, breaking the $Ξ©_m-Οƒ_8$ degeneracy at the Fisher information level. It further reduces the parameter MSE across our cosmological parameter prior by $1.45\times$ over the best classical mark. The learned latent geometry aligns with the $Ξ©_m$ and $Οƒ_8$ directions in parameter space, indicating that the contrastive objective recovers the dominant axes of cosmological information. Our approach opens the door to more powerful, interpretable summary statistics for cosmological inference.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

πŸ“œ Similar Papers

In the same crypt β€” astro-ph.CO

R.I.P. πŸ‘» Ghosted

Exhaustive Symbolic Regression

Deaglan J. Bartlett, Harry Desmond, Pedro G. Ferreira

astro-ph.CO πŸ› IEEE TEC πŸ“š 38 cites 3 years ago