Searching for Novel Chemistry in Exoplanetary Atmospheres using Machine Learning for Anomaly Detection

TL;DR

This paper explores the use of machine learning anomaly detection methods to identify exoplanets with unusual atmospheric compositions from spectroscopic data, aiding in the discovery of novel chemistry and potential biosignatures.

Contribution

It demonstrates the feasibility of applying Local Outlier Factor and One Class SVM for anomaly detection in synthetic exoplanet spectra with varying noise levels.

Findings

Both ML methods effectively identify anomalies in synthetic spectra.

Performance varies with noise levels, as shown by ROC curves.

The approach enables rapid screening of large exoplanet datasets.

Abstract

The next generation of telescopes will yield a substantial increase in the availability of high-resolution spectroscopic data for thousands of exoplanets. The sheer volume of data and number of planets to be analyzed greatly motivate the development of new, fast and efficient methods for flagging interesting planets for reobservation and detailed analysis. We advocate the application of machine learning (ML) techniques for anomaly (novelty) detection to exoplanet transit spectra, with the goal of identifying planets with unusual chemical composition and even searching for unknown biosignatures. We successfully demonstrate the feasibility of two popular anomaly detection methods (Local Outlier Factor and One Class Support Vector Machine) on a large public database of synthetic spectra. We consider several test cases, each with different levels of instrumental noise. In each case, we use ROC curves to quantify and compare the performance of the two ML techniques.