NOWCRETE · SEPTEMBER 2026

Cosmology and AI.

PhD candidate at the University of Crete / FORTH and CosmoStat · CEA Paris-Saclay. I study the large-scale structure of the Universe through weak gravitational lensing — combining simulation-based inference, deep learning, and higher-order statistics.

Andreas Tersenov
NowPhD candidate · FORTH
VisitingCosmoStat · CEA Saclay
Based inHeraklion · Saclay
IAffiliations & collaborations
FORTHInstitute of Computer Science · Astrophysics
CEA SaclayCosmoStat lab
TITANEU ERA Chair · AI in Astrophysics
EuclidCosmological Survey
UNIONSCosmological Survey
II

News

Updates from the lab
Nov 24, 2025🏆 Our team came 2nd place at the NeurIPS 2025 Weak Lensing Uncertainty Challenge — an international competition on uncertainty-aware and OOD-detection AI methods for cosmology.
Nov 10, 2025Lecturing at the Sharjah Winter School for AstroStatistics on Bayesian inference, advanced sampling, and simulation-based inference.
Oct 06, 2025New Euclid HOWLS paper 🌌 — “Towards a DR1 application of higher-order weak-lensing statistics” is on arXiv. We show that higher-order statistics constrain dark energy 2.5× better than standard methods in a realistic Euclid DR1 setup.
Mar 31, 2025Our paper “Impact of mass-mapping algorithms on cosmology inference” has been published in Astronomy & Astrophysics 🎉
Jan 08, 2025Spending the next 3 months at the CosmoStat Lab, CEA Paris-Saclay.
III

Selected publications

View all →

Impact of mass mapping algorithms on cosmology inference

A. Tersenov, L. Baumont, M. Kilbinger, J.-L. Starck

A&A 698 (2025): A25 · arXiv:2501.06961 · 2025 · mass mapping · inference

PUBLISHEDarXiv ↗BibTeX

Mitigating Baryonic Effects in Weak Lensing with Higher-Order Statistics

A. Tersenov, S. Guerrini, J.-L. Starck, M. Kilbinger

Submitted to Astronomy & Astrophysics · 2026 · HOS · baryons

NEWarXiv ↗BibTeX

Euclid preparation: Towards a DR1 application of higher-order weak lensing statistics

Euclid HOWLS Collaboration · A. Tersenov et al.

A&A 707 (2026): A235 · arXiv:2510.04953 · 2026 · Euclid · HOWLS

arXiv ↗BibTeX

Want to collaborate, or just chat about cosmology and AI?

I'm open to research collaborations, talks, and conversations with industry teams working on probabilistic ML, scientific inference, and large-scale data.