# Peeking inside the Black Box: Interpreting Deep Learning Models for Exoplanet Atmospheric Retrievals

arXiv: https://arxiv.org/abs/2011.11284 · HTML: https://mlsanity.com/paper/2011.11284/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Nov 2020.

## Built on by (most cited first, 9 of 9)

- [Analytical Modelling of Exoplanet Transit Specroscopy with Dimensional Analysis and Symbolic Regression](https://mlsanity.com/paper/2112.11600/) (2021)
- [Unsupervised Machine Learning for Exploratory Data Analysis of Exoplanet Transmission Spectra](https://mlsanity.com/paper/2201.02696/) (2022)
- [Searching for Novel Chemistry in Exoplanetary Atmospheres using Machine Learning for Anomaly Detection](https://mlsanity.com/paper/2308.07604/) (2023)
- [Flow Matching for Atmospheric Retrieval of Exoplanets: Where Reliability meets Adaptive Noise Levels](https://mlsanity.com/paper/2410.21477/) (2024)
- [Reconstructing Atmospheric Parameters of Exoplanets Using Deep Learning](https://mlsanity.com/paper/2310.01227/) (2023)
- [Predicting Exoplanetary Features with a Residual Model for Uniform and Gaussian Distributions](https://mlsanity.com/paper/2406.10771/) (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)
- [Life, Machine Learning, and the Search for Habitability: Predicting Biosignature Fluxes for the Habitable Worlds Observatory](https://mlsanity.com/paper/2601.12557/) (2026)
- [Amortized Probabilistic Retrieval of Atmospheric CO2 from OCO-2 Spectra Using Deep Learning with Laplace Approximations and Normalizing Flows](https://mlsanity.com/paper/2606.17413/) (2026)

## Abstract (arXiv)

Deep learning algorithms are growing in popularity in the field of exoplanetary science due to their ability to model highly non-linear relations and solve interesting problems in a data-driven manner. Several works have attempted to perform fast retrievals of atmospheric parameters with the use of machine learning algorithms like deep neural networks (DNNs). Yet, despite their high predictive power, DNNs are also infamous for being 'black boxes'. It is their apparent lack of explainability that makes the astrophysics community reluctant to adopt them. What are their predictions based on? How confident should we be in them? When are they wrong and how wrong can they be? In this work, we present a number of general evaluation methodologies that can be applied to any trained model and answer questions like these. In particular, we train three different popular DNN architectures to retrieve atmospheric parameters from exoplanet spectra and show that all three achieve good predictive performance. We then present an extensive analysis of the predictions of DNNs, which can inform us - among other things - of the credibility limits for atmospheric parameters for a given instrument and model. Finally, we perform a perturbation-based sensitivity analysis to identify to which features of the spectrum the outcome of the retrieval is most sensitive. We conclude that for different molecules, the wavelength ranges to which the DNN's predictions are most sensitive, indeed coincide with their characteristic absorption regions. The methodologies presented in this work help to improve the evaluation of DNNs and to grant interpretability to their predictions.