Séminaire IPAG

---

Data-driven approaches for exploring circumstellar environments with high-contrast imaging

jeudi 2 octobre 2025 - 11h00

Olivier Flasseur - CRAL
--- 
The detection of exoplanets, the characterization of their atmospheres, and the study of exoplanet formation mechanisms are major current challenges in astrophysics. High-contrast direct imaging (HCI) is one of the observational techniques of choice to address these questions. However, such observations are particularly demanding due to the extreme contrast levels and angular resolution required. In addition to the use of extreme adaptive optics and coronagraphs, advances in data science have become critical for analyzing these observations and disentangling the signals of interest (exoplanets and circumstellar disks) from the strong nuisance component (speckles and noise) that corrupts the data. In this seminar, I will present recent advances in data science aimed at the optimal and reliable extraction of astrophysical information from multivariate observations. These approaches rely on accurate modeling of the different contributions to the total signal and of the multiple correlations that arise between them. Such a joint approach involves combining statistical and physical modeling, deep learning methods enriched with prior domain knowledge, and strategies for the fusion of heterogeneous data. Emphasis will be placed on (i) combining deep learning models with statistical modeling of the nuisance, (ii) leveraging large archival database as a valuable source of diversity for tackling the unmixing task, and (iii) jointly exploiting the spectral diversity of the observations. Using real data from high-contrast instruments (including VLT/SPHERE), I will illustrate that these approaches enable accurate modeling and effective subtraction of the nuisance component, leading to reliable and nearly-optimal estimates of the quantities of interest. This results in significantly improved detection sensitivity and reconstruction fidelity. Such a framework is also scalable and readily applicable to large-scale surveys. Looking ahead, instruments on the next generation of thirty-meter-class telescopes will enable the exploration of the innermost environments of Sun-like stars at unprecedented contrast levels. Achieving the associated scientific goals will require addressing several data science challenges: (i) approaching the ultimate performance limits of the instruments through optimal signal extraction, (ii) capturing complex, spatially structured nuisance exhibiting strong variability, and (iii) building robust nuisance models, particularly in the vicinity of the host star. I will discuss these challenges in light of the methodological developments presented.

---

Hôtes : Mickael Bonnefoy

Salle Manuel Forestini, 414 rue de la piscine, 38400 Saint Martin d'Hères