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

arXiv: https://arxiv.org/abs/2308.07604 · HTML: https://mlsanity.com/paper/2308.07604/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 57 references that are in the map.
- First posted on arXiv 15 Aug 2023.

## Builds on (most cited first, 8 of 8)

- [Analytical Modelling of Exoplanet Transit Specroscopy with Dimensional Analysis and Symbolic Regression](https://mlsanity.com/paper/2112.11600/) (2021)
- [An Ensemble of Bayesian Neural Networks for Exoplanetary Atmospheric Retrieval](https://mlsanity.com/paper/1905.10659/) (2019)
- [Peeking inside the Black Box: Interpreting Deep Learning Models for Exoplanet Atmospheric Retrievals](https://mlsanity.com/paper/2011.11284/) (2020)
- [Integrating Machine Learning for Planetary Science: Perspectives for the Next Decade](https://mlsanity.com/paper/2007.15129/) (2020)
- [Exoplanet Characterization using Conditional Invertible Neural Networks](https://mlsanity.com/paper/2202.00027/) (2022)
- [Convolutional neural networks as an alternative to Bayesian retrievals](https://mlsanity.com/paper/2203.01236/) (2022)
- [Incorporating Physical Knowledge into Machine Learning for Planetary Space Physics](https://mlsanity.com/paper/2006.01927/) (2020)
- [Postulating Exoplanetary Habitability via a Novel Anomaly Detection Method](https://mlsanity.com/paper/2109.02273/) (2021)

## Built on by (most cited first, 3 of 3)

- [Operational range bounding of spectroscopy models with anomaly detection](https://mlsanity.com/paper/2408.02581/) (2024)
- [Hunting for "Oddballs" with Machine Learning: Detecting Anomalous Exoplanets Using a Deep-Learned Low-Dimensional Representation of Transit Spectra with Autoencoders](https://mlsanity.com/paper/2601.02324/) (2026)
- [ASTER -- Agentic Science Toolkit for Exoplanet Research](https://mlsanity.com/paper/2603.26953/) (2026)

## Abstract (arXiv)

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.