PGNets: Planet mass prediction using convolutional neural networks for radio continuum observations of protoplanetary disks

Abstract

We developed Convolutional Neural Networks (CNNs) to rapidly and directly infer the planet mass from radio dust continuum images. Substructures induced by young planets in protoplanetary disks can be used to infer the potential young planets' properties. Hydrodynamical simulations have been used to study the relationships between the planet's properties and these disk features. However, these attempts either fine-tuned numerical simulations to fit one protoplanetary disk at a time, which was time-consuming, or azimuthally averaged simulation results to derive some linear relationships between the gap width/depth and the planet mass, which lost information on asymmetric features in disks. To cope with these disadvantages, we developed Planet Gap neural Networks (PGNets) to infer the planet mass from 2D images. We first fit the gridded data in Zhang et al. (2018) as a classification problem. Then, we quadrupled the data set by running additional simulations with near-randomly sampled parameters, and derived the planet mass and disk viscosity together as a regression problem. The classification approach can reach an accuracy of 92\%, whereas the regression approach can reach 1$\sigma$ as 0.16 dex for planet mass and 0.23 dex for disk viscosity. We can reproduce the degeneracy scaling $\alpha$ $\propto$ $M_p^3$ found in the linear fitting method, which means that the CNN method can even be used to find degeneracy relationship. The gradient-weighted class activation mapping effectively confirms that PGNets use proper disk features to constrain the planet mass. We provide programs for PGNets and the traditional fitting method from Zhang et al. (2018), and discuss each method's advantages and disadvantages.

arXiv comment12 pages, 7 figures, accepted to MNRAS

References in corpus (68)

• SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
• Astropy: A Community Python Package for Astronomy
• The Disk Substructures at High Angular Resolution Project (DSHARP): I. Motivation, Sample, Calibration, and Overview
• Rotation-invariant convolutional neural networks for galaxy morphology prediction
• Discovery of a planetary-mass companion within the gap of the transition disk around PDS 70
• Gaps and Rings in an ALMA Survey of Disks in the Taurus Star-forming Region
• The Disk Substructures at High Angular Resolution Project (DSHARP): II. Characteristics of Annular Substructures
• Two accreting protoplanets around the young star PDS 70
• Dust and Gas in the disc of HL Tauri: Surface density, dust settling, and dust-to-gas ratio
• The Disk Substructures at High Angular Resolution Project (DSHARP): VII. The Planet-Disk Interactions Interpretation
• The Disk Substructures at High Angular Resolution Project (DSHARP): V. Interpreting ALMA maps of protoplanetary disks in terms of a dust model
• A Kinematical Detection of Two Embedded Jupiter Mass Planets in HD 163296
• Kinematic evidence for an embedded protoplanet in a circumstellar disc
• Observational Signatures of Planets in Protoplanetary Disks I: Gaps Opened by Single and Multiple Young Planets in Disks
• On planet formation in HL Tau
• Fast Automated Analysis of Strong Gravitational Lenses with Convolutional Neural Networks
• Orbital and atmospheric characterization of the planet within the gap of the PDS 70 transition disk
• On the corotation torque in a radiatively inefficient disk
• How Empty are Disk Gaps Opened by Giant Planets?
• Turbulence in the TW Hya Disk
• Dust flow in gas disks in the presence of embedded planets
• Mass Estimates of a Giant Planet in a Protoplanetary Disk from the Gap Structures
• One Solution to the Mass Budget Problem for Planet Formation: Optically Thick Disks with Dust Scattering
• The minimum mass of detectable planets in protoplanetary discs and the derivation of planetary masses from high resolution observations
• Submillimeter emission associated with candidate protoplanets
• Magellan Adaptive Optics Imaging of PDS 70: Measuring the Mass Accretion Rate of a Young Giant Planet within a Gapped Disk
• Type I planetary migration in a self-gravitating disk
• Particle Concentration At Planet Induced Gap Edges and Vortices: I. Inviscid 3-D Hydro Disks
• The newborn planet population emerging from ring-like structures in discs
• Mass constraint for a planet in a protoplanetary disk from the gap width
• The anomalously low (sub)millimeter spectral indices of some protoplanetary disks may be explained by dust self-scattering
• Planet formation signposts: observability of circumplanetary disks via gas kinematics
• The Ophiuchus DIsc Survey Employing ALMA (ODISEA)-III: the evolution of substructures in massive discs at 3-5 au resolution
• Multiple Disk Gaps and Rings Generated by a Single Super-Earth: II. Spacings, Depths, and Number of Gaps, with Application to Real Systems
• Planet-driven spiral arms in protoplanetary disks: I. Formation mechanism
• Rings and gaps in the disc around Elias 24 revealed by ALMA
• Gas and dust hydrodynamical simulations of massive lopsided transition discs - II. Dust concentration
• An opening criterion for dust gaps in protoplanetary discs
• Keck/NIRC2 $L$'-Band Imaging of Jovian-Mass Accreting Protoplanets around PDS 70
• Accretion Properties of PDS 70b with MUSE
• Planet-disk interaction in disks with cooling: basic theory
• Planet-driven spiral arms in protoplanetary disks: II. Implications
• Disk Substructures at High Angular Resolution Program (DSHARP): VIII. The Rich Ringed Substructures in the AS 209 Disk
• The Disc Miner I: A statistical framework to detect and quantify kinematical perturbations driven by young planets in discs
• The widest H$\alpha$ survey of accreting protoplanets around nearby transition disks
• Separating extended disc features from the protoplanet in PDS 70 using VLT/SINFONI
• How do giant planetary cores shape the dust disk? HL Tau system
• Identifying Reionization Sources from 21cm Maps using Convolutional Neural Networks
• The effects of disk self-gravity and radiative cooling on the formation of gaps and spirals by young planets
• Deep imaging search for planets forming in the TW Hya protoplanetary disk with the Keck/NIRC2 vortex coronagraph
• Impacts of dust feedback on a dust ring induced by a planet in a protoplanetary disk
• Gaps and Rings in Protoplanetary Disks with Realistic Thermodynamics: The Critical Role of In-Plane Radiation Transport
• Detecting solar system objects with convolutional neural networks
• A super-resolution analysis of the DSHARP survey: Substructure is common in the inner 30 au
• Revealing signatures of planets migrating in protoplanetary discs with ALMA multi-wavelength observations
• Constraining Protoplanetary Disk Accretion and Young Planets Using ALMA Kinematic Observations
• Importance of radiative effects in gap opening by planets in protoplanetary disks
• CASI: A Convolutional Neural Network Approach for Shell Identification
• Meso-scale Instability Triggered by Dust Feedback in Dusty Rings: Origin and Observational Implications
• Using Deep Neural Networks to compute the mass of forming planets
• Application of Convolutional Neural Networks to Identify Stellar Feedback Bubbles in CO Emission
• Morphological signatures induced by dust back reaction in discs with an embedded planet
• Migrating Low-Mass Planets in Inviscid Dusty Protoplanetary Discs
• Application of Convolutional Neural Networks to Identify Protostellar Outflows in CO Emission
• Hiding Signatures of Gravitational Instability in Protoplanetary Discs with Planets
• 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
• Model of a gap formed by a planet with fast inward migration

Cited by in corpus (8)

• PPDONet: Deep Operator Networks for Fast Prediction of Steady-State Solutions in Disk-Planet Systems
• Substructures in Compact Disks of the Taurus Star-forming Region
• Denoising Diffusion Probabilistic Models to Predict the Density of Molecular Clouds
• DBNets: A publicly available deep learning tool to measure the masses of young planets in dusty protoplanetary discs
• Using Bayesian Deep Learning to infer Planet Mass from Gaps in Protoplanetary Disks
• DBNets2.0: simulation-based inference for planet-induced dust substructures in protoplanetary discs
• Kinematic Evidence of an Embedded Protoplanet in HD 142666 Identified by Machine Learning
• exoALMA XXIII. Estimating Disk and Planet Properties from Dust Morphologies with DBNets2.0