DBNets: A publicly available deep learning tool to measure the masses of young planets in dusty protoplanetary discs

Abstract

Current methods to characterize embedded planets in protoplanetary disc observations are severely limited either in their ability to fully account for the observed complex physics or in their computational and time costs. To address this shortcoming, we developed DBNets: a deep learning tool, based on convolutional neural networks, that analyses substructures observed in the dust continuum emission of protoplanetary discs to quickly infer the mass of allegedly embedded planets. We focussed on developing a method to reliably quantify not only the planet mass, but also the associated uncertainty introduced by our modelling and adopted techniques. Our tests gave promising results achieving an 87% reduction of the log Mp mean squared error with respect to an analytical formula fitted on the same data (DBNets metrics: lmse 0.016, r2-score 97%). With the goal of providing the final user of DBNets with all the tools needed to interpret their measurements and decide on their significance, we extensively tested our tool on out-of-distribution data. We found that DBNets can identify inputs strongly outside its training scope returning an uncertainty above a specific threshold and we thus provided a rejection criterion that helps determine the significance of the results obtained. Additionally, we outlined some limitations of our tool: it can be reliably applied only on discs observed with inclinations below approximately 60°, in the optically thin regime, with a resolution 8 times better than the gap radial location and with a signal-to-noise ratio higher than approximately ten. Finally, we applied DBNets to 33 actual observations of protoplanetary discs measuring the mass of 48 proposed planets and comparing our results with the available literature. We confirmed that most of the observed gaps imply planets in the sub-Jupiter regime. DBNets is publicly available at dbnets.fisica.unimi.it.

References in corpus (39)

• The Gemini Planet Imager Exoplanet Survey: Giant Planet and Brown Dwarf Demographics From 10-100 AU
• A comparative study of disc-planet interaction
• Evidence of fast pebble growth near condensation fronts in the HL Tau protoplanetary disk
• Discovery of a Companion Candidate in the HD169142 Transition Disk and the Possibility of Multiple Planet Formation
• Nine localised deviations from Keplerian rotation in the DSHARP circumstellar disks: Kinematic evidence for protoplanets carving the gaps
• Images of Embedded Jovian Planet Formation At A Wide Separation Around AB Aurigae
• The newborn planet population emerging from ring-like structures in discs
• On the formation of multiple concentric rings and gaps in protoplanetary disks
• Dust Density Distribution and Imaging Analysis of Different Ice Lines in Protoplanetary Disks
• The Ophiuchus DIsc Survey Employing ALMA (ODISEA)-III: the evolution of substructures in massive discs at 3-5 au resolution
• Dust unveils the formation of a mini-Neptune planet in a protoplanetary ring
• An opening criterion for dust gaps in protoplanetary discs
• Dust-driven viscous ring-instability in protoplanetary disks
• Perturbers: SPHERE detection limits to planetary-mass companions in protoplanetary disks
• A new planet candidate detected in a dust gap of the disc around HD 163296 through localised kinematic signatures. An observational validation of the Discminer
• Annular substructures in the transition disks around LkCa 15 and J1610
• Dust traps in the protoplanetary disc MWC 758: two vortices produced by two giant planets?
• An Ensemble of Bayesian Neural Networks for Exoplanetary Atmospheric Retrieval
• Blobs, spiral arms, and a possible planet around HD 169142
• Predicting the observational signature of migrating Neptune-sized planets in low-viscosity disks
• The effects of disk self-gravity and radiative cooling on the formation of gaps and spirals by young planets
• Large-scale CO spiral arms and complex kinematics associated with the T Tauri star RU Lup
• Revealing signatures of planets migrating in protoplanetary discs with ALMA multi-wavelength observations
• Long lived dust rings around HD169142
• Constraining planetesimal stirring: how sharp are debris disc edges?
• ALMA Observations of the Asymmetric Dust Disk around DM Tau
• Mapping the Planetary Wake in HD 163296 with Kinematics
• PPDONet: Deep Operator Networks for Fast Prediction of Steady-State Solutions in Disk-Planet Systems
• Gap and rings carved by vortices in protoplanetary dust
• Analysing the SEDs of protoplanetary disks with machine learning
• The detection of dust gap-ring structure in the outer region of the CR Cha protoplanetary disk
• PGNets: Planet mass prediction using convolutional neural networks for radio continuum observations of protoplanetary disks
• A Machine Learning model to infer planet masses from gaps observed in protoplanetary disks
• DPNNet-2.0 Part I: Finding hidden planets from simulated images of protoplanetary disk gaps
• Architecture of planetary systems predicted from protoplanetary disks observed with ALMA I: mass of the possible planets embedded in the dust gap
• Do all gaps in protoplanetary discs host planets?
• Locating Hidden Exoplanets in ALMA Data Using Machine Learning
• Substructures in protoplanetary disks imprinted by compact planetary systems
• The Magnetic Field versus Density relation in Star-Forming Molecular Clouds

Cited by in corpus (8)

• exoALMA IV: Substructures, Asymmetries, and the Faint Outer Disk in Continuum Emission
• A high-resolution survey of protoplanetary disks in Lupus and the nature of compact disks
• A young gas giant and hidden substructures in a protoplanetary disk
• DBNets2.0: simulation-based inference for planet-induced dust substructures in protoplanetary discs
• Multi-frequency analysis of the ALMA and VLA high resolution continuum observations of the substructured disc around CI Tau. Preference for sub-mm-sized low-porosity amorphous carbon grains
• Hidden under a warm blanket: If planets existed in protostellar disks, they would hardly produce observable substructures
• Protoplanetary disk insights from the first ERIS/APP survey at 4 μm
• exoALMA XXIII. Estimating Disk and Planet Properties from Dust Morphologies with DBNets2.0