Identifying Exoplanets with Deep Learning: A Five Planet Resonant Chain around Kepler-80 and an Eighth Planet around Kepler-90

Abstract

NASA's Kepler Space Telescope was designed to determine the frequency of Earth-sized planets orbiting Sun-like stars, but these planets are on the very edge of the mission's detection sensitivity. Accurately determining the occurrence rate of these planets will require automatically and accurately assessing the likelihood that individual candidates are indeed planets, even at low signal-to-noise ratios. We present a method for classifying potential planet signals using deep learning, a class of machine learning algorithms that have recently become state-of-the-art in a wide variety of tasks. We train a deep convolutional neural network to predict whether a given signal is a transiting exoplanet or a false positive caused by astrophysical or instrumental phenomena. Our model is highly effective at ranking individual candidates by the likelihood that they are indeed planets: 98.8% of the time it ranks plausible planet signals higher than false positive signals in our test set. We apply our model to a new set of candidate signals that we identified in a search of known Kepler multi-planet systems. We statistically validate two new planets that are identified with high confidence by our model. One of these planets is part of a five-planet resonant chain around Kepler-80, with an orbital period closely matching the prediction by three-body Laplace relations. The other planet orbits Kepler-90, a star which was previously known to host seven transiting planets. Our discovery of an eighth planet brings Kepler-90 into a tie with our Sun as the star known to host the most planets.

Accepted for publication in the Astronomical Journal. 23 pages, 12 figures, 6 tables

References in corpus (20)

• Seven temperate terrestrial planets around the nearby ultracool dwarf star TRAPPIST-1
• Most 1.6 Earth-Radius Planets are not Rocky
• A Technique for Extracting Highly Precise Photometry for the Two-Wheeled Kepler Mission
• KEPLER's First Rocky Planet: Kepler-10b
• A stellar-mass-dependent drop in planet occurrence rates
• Exoplanet population inference and the abundance of Earth analogs from noisy, incomplete catalogs
• Planetary Candidates Observed by Kepler VI: Planet Sample from Q1-Q16 (47 Months)
• The Kepler Follow-Up Observation Program. I. A Catalog of Companions to Kepler Stars from High-Resolution Imaging
• A Study of the Shortest-Period Planets Found With Kepler
• The Mass of Kepler-93b and The Composition of Terrestrial Planets
• A resonant chain of four transiting, sub-Neptune planets
• Planetary Candidates Observed by Kepler V: Planet Sample from Q1-Q12 (36 Months)
• A Dynamical Analysis of the Kepler-80 System of Five Transiting Planets
• Deep-HiTS: Rotation Invariant Convolutional Neural Network for Transient Detection
• Automatic Classification of Kepler Planetary Transit Candidates
• The population of long-period transiting exoplanets
• Detection of Stars within $\sim$0.8 arcseconds of Kepler Objects of Interest
• Transit Shapes and Self Organising Maps as a Tool for Ranking Planetary Candidates: Application to Kepler and K2
• On the Inclination and Habitability of the HD 10180 System
• Supervised Learning Detection of Sixty Non-Transiting Hot Jupiter Candidates

Cited by in corpus (147)

• The Challenge of Machine Learning in Space Weather Nowcasting and Forecasting
• The TESS Objects of Interest Catalog from the TESS Prime Mission
• Transit Least Squares: Optimized transit detection algorithm to search for periodic transits of small planets
• Exoplanet Statistics and Theoretical Implications
• Lunar Crater Identification via Deep Learning
• Vetting of 384 TESS Objects of Interest with TRICERATOPS and Statistical Validation of 12 Planet Candidates
• Flare Statistics for Young Stars from a Convolutional Neural Network Analysis of $\text{TESS}$ Data
• Surveying the reach and maturity of machine learning and artificial intelligence in astronomy
• Radio Galaxy Zoo: ClaRAN - A Deep Learning Classifier for Radio Morphologies
• Accounting for Incompleteness due to Transit Multiplicity in Kepler Planet Occurrence Rates
• ExoGAN: Retrieving Exoplanetary Atmospheres Using Deep Convolutional Generative Adversarial Networks
• Scientific Domain Knowledge Improves Exoplanet Transit Classification with Deep Learning
• 44 Validated Planets from K2 Campaign 10
• Learning physical properties of liquid crystals with deep convolutional neural networks
• ExoMiner: A Highly Accurate and Explainable Deep Learning Classifier that Validates 301 New Exoplanets
• Identifying Exoplanets with Deep Learning III: Automated Triage and Vetting of TESS Candidates
• Identifying Exoplanets with Deep Learning II: Two New Super-Earths Uncovered by a Neural Network in K2 Data
• An Ensemble of Bayesian Neural Networks for Exoplanetary Atmospheric Retrieval
• Machine-learning Approaches to Exoplanet Transit Detection and Candidate Validation in Wide-field Ground-based Surveys
• Discovery and Vetting of Exoplanets I: Benchmarking K2 Vetting Tools
• The Kepler Giant Planet Search. I: A Decade of Kepler Planet-host Radial Velocities from W. M. Keck Observatory
• An all-sky sample of intermediate- to high-mass OBA-type eclipsing binaries observed by TESS
• Shallow Transits - Deep Learning I: Feasibility Study of Deep Learning to Detect Periodic Transits of Exoplanets
• Rapid Classification of TESS Planet Candidates with Convolutional Neural Networks
• A Compact Multi-Planet System With A Significantly Misaligned Ultra Short Period Planet
• Scalable End-to-end Recurrent Neural Network for Variable star classification
• The SOPHIE search for northern extrasolar planets. XVI. HD 158259: A compact planetary system in a near-3:2 mean motion resonance chain
• Machine learning inference of the interior structure of low-mass exoplanets
• Deep Learning for Image Sequence Classification of Astronomical Events
• On Neural Architectures for Astronomical Time-series Classification with Application to Variable Stars
• A framework for the architecture of exoplanetary systems. I. Four classes of planetary system architecture
• Exoplanet Validation with Machine Learning: 50 new validated Kepler planets
• Warm Jupiters in TESS Full-Frame Images: A Catalog and Observed Eccentricity Distribution for Year 1
• TESS Delivers Five New Hot Giant Planets Orbiting Bright Stars from the Full Frame Images
• Assessment of Supervised Machine Learning for Atmospheric Retrieval of Exoplanets
• Giants are bullies: how their growth influences systems of inner sub-Neptunes and super-Earths
• The CARMENES search for exoplanets around M dwarfs -- A deep learning approach to determine fundamental parameters of target stars
• The TESS Grand Unified Hot Jupiter Survey. II. Twenty New Giant Planets
• Detecting Solar-like Oscillations in Red Giants with Deep Learning
• Classifying Exoplanet Candidates with Convolutional Neural Networks: Application to the Next Generation Transit Survey
• Formation of compact systems of super-Earths via dynamical instabilities and giant impacts
• How Jupiters save or destroy inner Neptunes around evolved stars
• AutoRegressive Planet Search: Application to the Kepler Mission
• Validation of 13 Hot and Potentially Terrestrial TESS Planets
• Quantum topology identification with deep neural networks and quantum walks
• Into the Darkness: Classical and Type II Cepheids in the Zona Galactica Incognita
• A search for multi-planet systems with TESS using a Bayesian N-body retrieval and machine learning
• Kepler Object of Interest Network II. Photodynamical modelling of Kepler-9 over 8 years of transit observations
• Three Pathways for Observed Resonant Chains
• AutoRegressive Planet Search: Methodology
• Alleviating the transit timing variation bias in transit surveys. I. RIVERS: Method and detection of a pair of resonant super-Earths around Kepler-1705
• Deep Neural Network Classifier for Variable Stars with Novelty Detection Capability
• Another Shipment of Six Short-Period Giant Planets from TESS
• TESS Reveals HD 118203 b to be a Transiting Planet
• SILVERRUSH X: Machine Learning-Aided Selection of $9318$ LAEs at $z=2.2$, $3.3$, $4.9$, $5.7$, $6.6$, and $7.0$ from the HSC SSP and CHORUS Survey Data
• The TESS Grand Unified Hot Jupiter Survey. I. Ten TESS Planets
• Evolution of Novel Activation Functions in Neural Network Training with Applications to Classification of Exoplanets
• Predicting binding energies of astrochemically relevant molecules via machine learning
• A Machine-Learning-Based Direction-of-Origin Filter for the Identification of Radio Frequency Interference in the Search for Technosignatures
• Re-Evaluating Small Long-Period Confirmed Planets From Kepler
• Detecting Exoplanet Transits through Machine Learning Techniques with Convolutional Neural Networks
• Identifying Exoplanets with Deep Learning. V. Improved Light Curve Classification for TESS Full Frame Image Observations
• Automated identification of transiting exoplanet candidates in NASA Transiting Exoplanets Survey Satellite (TESS) data with machine learning methods
• Identification of BASS DR3 Sources as Stars, Galaxies and Quasars by XGBoost
• Deriving star cluster parameters with convolutional neural networks. I. Age, mass, and size
• Orbital Architectures of Kepler Multis From Dynamical Instabilities
• A Five-Planet Resonant Chain: Reevaluation of the Kepler-80 System
• Beyond the Local Volume. I. Surface Densities of Ultracool Dwarfs in Deep HST/WFC3 Parallel Fields
• Multiband galaxy morphologies for CLASH: a convolutional neural network transferred from CANDELS
• Classification of Periodic Variable Stars with Novel Cyclic-Permutation Invariant Neural Networks
• Searching for hot subdwarf stars from the LAMOST spectra. III. classifying the hot subdwarf stars from LAMOST DR4 using deep learning method
• A Machine Learning approach for correcting radial velocities using physical observables
• Detrending Exoplanetary Transit Light Curves with Long Short-Term Memory Networks
• The TESS Triple-9 Catalog: 999 uniformly vetted candidate exoplanets
• Multiplicity Boost Of Transit Signal Classifiers: Validation of 69 New Exoplanets Using The Multiplicity Boost of ExoMiner
• Updated Catalog of Kepler Planet Candidates: Focus on Accuracy and Orbital Periods
• Identify Light-Curve Signals with Deep Learning Based Object Detection Algorithm. I. Transit Detection
• PyLightcurve-torch: a transit modelling package for deep learning applications in PyTorch
• Application of Convolutional Neural Networks to Identify Protostellar Outflows in CO Emission
• Nigraha: Machine-learning based pipeline to identify and evaluate planet candidates from TESS
• Two Massive Jupiters in Eccentric Orbits from the TESS Full Frame Images
• The Weird Detector: Flagging periodic, coherent signals of arbitrary shape in time series photometry
• The GALAH Survey: Using Galactic Archaeology to Refine our Knowledge of TESS Target Stars
• Improving Earth-like planet detection in radial velocity using deep learning
• Morphological Parameters and Associated Uncertainties for 8 Million Galaxies in the Hyper Suprime-Cam Wide Survey
• Identifying Light-curve Signals with a Deep Learning Based Object Detection Algorithm. II. A General Light Curve Classification Framework
• GaMPEN: A Machine Learning Framework for Estimating Bayesian Posteriors of Galaxy Morphological Parameters
• A Pluto--Charon Sonata: Dynamical Limits on the Masses of the Small Satellites
• Matched filtering with non-Gaussian noise for planet transit detections
• On the Degree of Dynamical Packing in the Kepler Multi-planet Systems
• The TESS Triple-9 Catalog II: a new set of 999 uniformly-vetted exoplanet candidates
• X-ray Study of Spatial Structures in Tycho's Supernova Remnant Using Unsupervised Deep Learning
• The K2 & TESS Synergy II: Revisiting 26 systems in the TESS Primary Mission
• The Gliese 86 Binary System: A Warm Jupiter Formed in a Disk Truncated at $\approx$2 AU
• The Terrestrial Planet Formation around M Dwarfs: In-situ, Inward Migration or Reversed Migration
• Distinguishing a planetary transit from false positives: a Transformer-based classification for planetary transit signals
• One-dimensional Convolutional Neural Networks for Detecting Transiting Exoplanets
• AstroCatR: a Mechanism and Tool for Efficient Time Series Reconstruction of Large-Scale Astronomical Catalogues
• Neural network-based emulation of interstellar medium models
• Kepler data analysis: non-Gaussian noise and Fourier Gaussian process analysis of star variability
• Insightful analysis of historical sources at scales beyond human capabilities using unsupervised Machine Learning and XAI
• HD 83443c: A highly eccentric giant planet on a 22-year orbit
• A Census of Protostellar Outflows in Nearby Molecular Clouds
• Kepler-80 Revisited: Assessing the Participation of a Newly Discovered Planet in the Resonant Chain
• exoMMR: a New Python Package to Confirm and Characterize Mean Motion Resonances
• Estimating Planetary Mass with Deep Learning
• DIAmante TESS AutoRegressive Planet Search (DTARPS): I. Analysis of 0.9 Million Light Curves
• A Roadmap to Domain Knowledge Integration in Machine Learning
• Identifying Exoplanets with Deep Learning. IV. Removing Stellar Activity Signals from Radial Velocity Measurements Using Neural Networks
• Classification of Equation of State in Relativistic Heavy-Ion Collisions Using Deep Learning
• The GPU Phase Folding and Deep Learning Method for Detecting Exoplanet Transits
• Multi-Wavelength Photometry Derived from Monochromatic Kepler Data
• Searching for Possible Exoplanet Transits from BRITE Data through a Machine Learning Technique
• Identifying Potential Exomoon Signals with Convolutional Neural Networks
• Planet Occurrence: Doppler and Transit Surveys
• ExoplANNET: A deep learning algorithm to detect and identify planetary signals in radial velocity data
• DIAmante TESS AutoRegressive Planet Search (DTARPS): II. Hundreds of New TESS Candidate Exoplanets
• The reliability of the Titius-Bode relation and its implications for the search for exoplanets
• Using a neural network approach to accelerate disequilibrium chemistry calculations in exoplanet atmospheres
• Searching the Entirety of Kepler Data. I. 17 New Planet Candidates Including 1 Habitable Zone World
• Exoplanets Prediction in Multiplanetary Systems
• Computing Transiting Exoplanet Parameters with 1D Convolutional Neural Networks
• Searching for Sub-Second Stellar Variability with Wide-Field Star Trails and Deep Learning
• Classifying CMB time-ordered data through deep neural networks
• Disentangling stellar atmospheric parameters in astronomical spectra using Generative Adversarial Neural Networks
• CLOVER: Convnet Line-fitting Of Velocities in Emission-line Regions
• Discovery and Characterization of Kepler-36b
• Time delay estimation in unresolved lensed quasars
• Search for exocomets transits in Kepler light curves: Ten new transits identified
• Reducing false positives in strong lens detection through effective augmentation and ensemble learning
• Biases from Missing a Small Planet in High Multiplicity Systems
• Datacube segmentation via Deep Spectral Clustering
• DART-Vetter: A Deep LeARning Tool for automatic triage of exoplanet candidates
• PSF-based Analysis for Detecting Unresolved Wide Binaries
• Orbital stability of compact three-planet systems III. The role of three-body resonances
• $\text{Exoformer}$: Accelerating Bayesian atmospheric retrievals with transformer neural networks
• ExoDNN: Boosting exoplanet detection with artificial intelligence. Application to Gaia Data Release 3
• Accounting for Transit Timing Detectability: Biases in Planetary Radius and Orbital Period
• How to measure tidal dissipation in long resonant chains
• Radio Galaxy Zoo: Giant Radio Galaxy Classification using Multi-Domain Deep Learning
• Shallow Transits -- Deep Learning II: Identify Individual Exoplanetary Transits in Red Noise using Deep Learning
• Light curve fingerprints: an automated approach to the extraction of X-ray variability patterns with feature aggregation -- an example application to GRS 1915+105
• Panopticon: a novel deep learning model to detect single transit events with no prior data filtering in PLATO light curves
• Do the planets in the HD 34445 system really exist?
• The Dynamical Environment of Kepler-223 Planetary System
• Fourier series for eclipses on exoplanet binaries
• Exploring Long-period Architectures: Four New Planet Candidates from Kepler with Periods >342 days