Seminal
The magazine and atlas of EU-funded researchA new publication from the web and AI team at Scott Charnley

Atlas / Malta

Marie Skłodowska-Curie Action · Running

MALTA4COSMO MAchine Learning Techniques and Advanced statistics FOR COSMOlogy

Coordinator
Universita Ta Malta, Malta
Runs
7 September 2026 to 6 September 2028
Status
Running
EU funding
€205k
Total cost
—
Partners
1 organisations in 1 countries
Type
Marie Skłodowska-Curie Action
Grant agreement
101272532

What the project is doing

In the project's own words, from its summary on CORDIS.

Some of the most pressing open problems in modern cosmology concern the nature of the cosmic dark sector, the mismatch between the theoretically predicted and observed values of the cosmological constant, and the growing number of cosmological tensions between independent observational probes. Most notably, the Hubble constant shows a significant and persistent statistical discrepancy between local determinations and those inferred from early-universe Cosmic Microwave Background data.

These inconsistencies strongly suggest the need for new physics beyond the standard cosmological model. At the same time, current and forthcoming surveys are generating unprecedented volumes of data, whose accumulation and analysis demand substantial advances in statistical methodology before they can be effectively applied to targeted scientific questions. The MALTA4COSMO project aims to confront these challenges by developing a deep learning framework that incorporates advanced statistical and machine learning techniques, such as neural networks, into cosmological community codes for emulated parameter inference.

This approach will enable the identification, in a model-independent way, of generic features of cosmological theories that remain compatible with all available observations. Harnessing the power of deep learning thus provides an innovative arena to explore new physics in the observational sector and to discriminate between competing models that appear degenerate under current methods, advancing data-driven science as a crucial complement to theory-driven approaches.

The fellowship will support the researcher’s career development by fostering expertise in machine learning and Bayesian statistics, while also cultivating transversal skills in dissemination and outreach. Moreover, it will expand the researcher’s professional network within Europe and internationally, consolidating a long-term trajectory at the interface of cosmology, data science, and fundamental physics.

Topics

Who is involved

OrganisationCityCountryRole
Universita Ta MaltaMsidaMaltaCoordinator

Source: CORDIS, European Commission. Data of 5 Nov 2026, reused under the Commission's reuse policy. View on CORDIS · Grant DOI