Scientific Domain Knowledge Improves Exoplanet Transit Classification with Deep Learning

Abstract

Space-based missions such as Kepler, and soon TESS, provide large datasets that must be analyzed efficiently and systematically. Recent work by Shallue & Vanderburg (2018) successfully used state-of-the-art deep learning models to automatically classify Kepler transit signals as either exoplanets or false positives; our application of their model yielded 95.8% accuracy and 95.5% average precision. Here we expand upon that work by including additional scientific domain knowledge into the network architecture and input representations to significantly increase overall model performance to 97.5% accuracy and 98.0% average precision. Notably, we achieve 15-20% gains in recall for the lowest signal-to-noise transits that can correspond to rocky planets in the habitable zone. We input into the network centroid time-series information derived from Kepler data plus key stellar parameters taken from the Kepler DR25 catalogue. We also implement data augmentation techniques to alleviate model over-fitting. These improvements allow us to drastically reduce the size of the model, while still maintaining improved performance; smaller models are better for generalization, for example from Kepler to TESS data. This work illustrates the importance of including expert domain knowledge in even state-of-the-art deep learning models when applying them to scientific research problems that seek to identify weak signals in noisy data. This classification tool will be especially useful for upcoming space-based photometry missions focused on finding small planets, such as TESS and PLATO.

8 pages, 5 figures, accepted to ApJ Letters

References in corpus (18)

• Astropy: A Community Python Package for Astronomy
• Gaia Data Release 2. Summary of the contents and survey properties
• Binary companions of evolved stars in APOGEE DR14: Search method and catalog of ~5,000 companions
• The PLATO 2.0 Mission
• Planetary Candidates Observed by Kepler, III: Analysis of the First 16 Months of Data
• Characteristics of planetary candidates observed by Kepler, II: Analysis of the first four months of data
• Kepler Presearch Data Conditioning II - A Bayesian Approach to Systematic Error Correction
• Kepler Presearch Data Conditioning I - Architecture and Algorithms for Error Correction in Kepler Light Curves
• Overview of the Kepler Science Processing Pipeline
• A Technique for Extracting Highly Precise Photometry for the Two-Wheeled Kepler Mission
• Identifying Exoplanets with Deep Learning: A Five Planet Resonant Chain around Kepler-80 and an Eighth Planet around Kepler-90
• Revised Stellar Properties of Kepler Targets for the Q1-17 (DR25) Transit Detection Run
• Planetary Candidates Observed by Kepler IV: Planet Sample From Q1-Q8 (22 Months)
• They are Small Worlds After All: Revised Properties of Kepler M Dwarf Stars and their Planets
• 275 Candidates and 149 Validated Planets Orbiting Bright Stars in K2 Campaigns 0-10
• Searching for Exoplanets Using Artificial Intelligence
• Automatic Classification of Kepler Planetary Transit Candidates
• Shallow Transits - Deep Learning I: Feasibility Study of Deep Learning to Detect Periodic Transits of Exoplanets

Cited by in corpus (39)

• Explainable Machine Learning for Scientific Insights and Discoveries
• 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
• 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
• Discovery and Vetting of Exoplanets I: Benchmarking K2 Vetting Tools
• A Joint Mass-Radius-Period Distribution of Exoplanets
• Rapid Classification of TESS Planet Candidates with Convolutional Neural Networks
• On Neural Architectures for Astronomical Time-series Classification with Application to Variable Stars
• Exoplanet Validation with Machine Learning: 50 new validated Kepler planets
• Exocomets from a Solar System Perspective
• Classifying Exoplanet Candidates with Convolutional Neural Networks: Application to the Next Generation Transit Survey
• Validation of 13 Hot and Potentially Terrestrial TESS Planets
• Testing Self-Organized Criticality Across the Main Sequence using Stellar Flares from TESS
• Identifying Exoplanets with Deep Learning. V. Improved Light Curve Classification for TESS Full Frame Image Observations
• Detrending Exoplanetary Transit Light Curves with Long Short-Term Memory Networks
• Identify Light-Curve Signals with Deep Learning Based Object Detection Algorithm. I. Transit Detection
• Multiplicity Boost Of Transit Signal Classifiers: Validation of 69 New Exoplanets Using The Multiplicity Boost of ExoMiner
• Nigraha: Machine-learning based pipeline to identify and evaluate planet candidates from TESS
• Half-sibling regression meets exoplanet imaging: PSF modeling and subtraction using a flexible, domain knowledge-driven, causal framework
• An Interpretable Machine Learning Framework for Modeling High-Resolution Spectroscopic Data
• Assessment of astronomical images using combined machine learning models
• One-dimensional Convolutional Neural Networks for Detecting Transiting Exoplanets
• Distinguishing a planetary transit from false positives: a Transformer-based classification for planetary transit signals
• A Roadmap to Domain Knowledge Integration in Machine Learning
• DIAmante TESS AutoRegressive Planet Search (DTARPS): I. Analysis of 0.9 Million Light Curves
• Identifying Exoplanets with Deep Learning. IV. Removing Stellar Activity Signals from Radial Velocity Measurements Using Neural Networks
• Searching for Possible Exoplanet Transits from BRITE Data through a Machine Learning Technique
• ExoplANNET: A deep learning algorithm to detect and identify planetary signals in radial velocity data
• Hunting for exocomet transits in the TESS database using the Random Forest method
• Following the TraCS of exoplanets with Pan-Planets: Wendelstein-1b and Wendelstein-2b
• Computing Transiting Exoplanet Parameters with 1D Convolutional Neural Networks
• DART-Vetter: A Deep LeARning Tool for automatic triage of exoplanet candidates
• Shallow Transits -- Deep Learning II: Identify Individual Exoplanetary Transits in Red Noise using Deep Learning
• ExoDNN: Boosting exoplanet detection with artificial intelligence. Application to Gaia Data Release 3
• Radio Galaxy Zoo: Giant Radio Galaxy Classification using Multi-Domain Deep Learning
• Panopticon: a novel deep learning model to detect single transit events with no prior data filtering in PLATO light curves